Star Hotels Project

Context

A significant number of hotel bookings are called-off due to cancellations or no-shows. The typical reasons for cancellations include change of plans, scheduling conflicts, etc. This is often made easier by the option to do so free of charge or preferably at a low cost which is beneficial to hotel guests but it is a less desirable and possibly revenue-diminishing factor for hotels to deal with. Such losses are particularly high on last-minute cancellations.

The new technologies involving online booking channels have dramatically changed customers’ booking possibilities and behavior. This adds a further dimension to the challenge of how hotels handle cancellations, which are no longer limited to traditional booking and guest characteristics.

The cancellation of bookings impact a hotel on various fronts:

• Loss of resources (revenue) when the hotel cannot resell the room.
• Additional costs of distribution channels by increasing commissions or paying for publicity to help sell these rooms.
• Lowering prices last minute, so the hotel can resell a room, resulting in reducing the profit margin.
• Human resources to make arrangements for the guests.

Objective

The increasing number of cancellations calls for a Machine Learning based solution that can help in predicting which booking is likely to be canceled. Star Hotels Group has a chain of hotels in Portugal, they are facing problems with the high number of booking cancellations and have reached out to your firm for data-driven solutions. You as a data scientist have to analyze the data provided to find which factors have a high influence on booking cancellations, build a predictive model that can predict which booking is going to be canceled in advance, and help in formulating profitable policies for cancellations and refunds.

Data Description

The data contains the different attributes of customers' booking details. The detailed data dictionary is given below.

Data Dictionary

• no_of_adults: Number of adults
• no_of_children: Number of Children
• no_of_weekend_nights: Number of weekend nights (Saturday or Sunday) the guest stayed or booked to stay at the hotel
• no_of_week_nights: Number of week nights (Monday to Friday) the guest stayed or booked to stay at the hotel
• type_of_meal_plan: Type of meal plan booked by the customer:
  • Not Selected – No meal plan selected
  • Meal Plan 1 – Breakfast
  • Meal Plan 2 – Half board (breakfast and one other meal)
  • Meal Plan 3 – Full board (breakfast, lunch, and dinner)
• required_car_parking_space: Does the customer require a car parking space? (0 - No, 1- Yes)
• room_type_reserved: Type of room reserved by the customer. The values are ciphered (encoded) by Star Hotels.
• lead_time: Number of days between the date of booking and the arrival date
• arrival_year: Year of arrival date
• arrival_month: Month of arrival date
• arrival_date: Date of the month
• market_segment_type: Market segment designation.
• repeated_guest: Is the customer a repeated guest? (0 - No, 1- Yes)
• no_of_previous_cancellations: Number of previous bookings that were canceled by the customer prior to the current booking
• no_of_previous_bookings_not_canceled: Number of previous bookings not canceled by the customer prior to the current booking
• avg_price_per_room: Average price per day of the reservation; prices of the rooms are dynamic. (in euros)
• no_of_special_requests: Total number of special requests made by the customer (e.g. high floor, view from the room, etc)
• booking_status: Flag indicating if the booking was canceled or not.

Importing necessary libraries and data

# this will help in making the Python code more structured automatically (good coding practice)
%load_ext nb_black

# Library to suppress warnings or deprecation notes
import warnings

warnings.filterwarnings("ignore")

# Libraries to help with reading and manipulating data

import pandas as pd
import numpy as np

# Library to split data
from sklearn.model_selection import train_test_split

# libaries to help with data visualization
import matplotlib.pyplot as plt
import seaborn as sns

# Removes the limit for the number of displayed columns
pd.set_option("display.max_columns", None)
# Sets the limit for the number of displayed rows
pd.set_option("display.max_rows", 200)

# To build model for prediction
import statsmodels.stats.api as sms
from statsmodels.stats.outliers_influence import variance_inflation_factor
import statsmodels.api as sm
from statsmodels.tools.tools import add_constant
from sklearn.linear_model import LogisticRegression


# Libraries to build decision tree classifier
from sklearn.tree import DecisionTreeClassifier
from sklearn import tree

# To tune different models
from sklearn.model_selection import GridSearchCV

# To get diferent metric scores
from sklearn.metrics import (
    f1_score,
    accuracy_score,
    recall_score,
    precision_score,
    confusion_matrix,
    roc_auc_score,
    plot_confusion_matrix,
    precision_recall_curve,
    roc_curve,
    make_scorer,
)

The nb_black extension is already loaded. To reload it, use:
  %reload_ext nb_black

Load the dataset

hotels = pd.read_csv("StarHotelsGroup.csv")

# copying data to another varaible to avoid any changes to original data
data = hotels.copy()

Data Overview

View the first and last 5 rows of the dataset.

data.head()

	no_of_adults	no_of_children	no_of_weekend_nights	no_of_week_nights	type_of_meal_plan	required_car_parking_space	room_type_reserved	lead_time	arrival_year	arrival_month	arrival_date	market_segment_type	repeated_guest	no_of_previous_cancellations	no_of_previous_bookings_not_canceled	avg_price_per_room	no_of_special_requests	booking_status
0	2	0	1	2	Meal Plan 1	0	Room_Type 1	224	2017	10	2	Offline	0	0	0	65.00	0	Not_Canceled
1	2	0	2	3	Not Selected	0	Room_Type 1	5	2018	11	6	Online	0	0	0	106.68	1	Not_Canceled
2	1	0	2	1	Meal Plan 1	0	Room_Type 1	1	2018	2	28	Online	0	0	0	60.00	0	Canceled
3	2	0	0	2	Meal Plan 1	0	Room_Type 1	211	2018	5	20	Online	0	0	0	100.00	0	Canceled
4	3	0	0	3	Not Selected	0	Room_Type 1	277	2019	7	13	Online	0	0	0	89.10	2	Canceled

data.tail()

	no_of_adults	no_of_children	no_of_weekend_nights	no_of_week_nights	type_of_meal_plan	required_car_parking_space	room_type_reserved	lead_time	arrival_year	arrival_month	arrival_date	market_segment_type	repeated_guest	no_of_previous_cancellations	no_of_previous_bookings_not_canceled	avg_price_per_room	no_of_special_requests	booking_status
56921	2	1	0	1	Meal Plan 2	0	Room_Type 4	45	2019	6	15	Online	0	0	0	163.88	1	Not_Canceled
56922	2	0	1	1	Meal Plan 1	0	Room_Type 1	320	2019	5	15	Offline	0	0	0	90.00	1	Canceled
56923	2	0	0	3	Not Selected	0	Room_Type 1	63	2018	4	21	Online	0	0	0	94.50	0	Canceled
56924	2	0	2	2	Not Selected	0	Room_Type 1	6	2019	4	28	Online	0	0	0	162.50	2	Not_Canceled
56925	2	0	1	2	Meal Plan 1	0	Room_Type 1	207	2018	12	30	Offline	0	0	0	161.67	0	Not_Canceled

Understand the shape of the dataset.

data.shape

(56926, 18)

• Dataset has 56926 rows and 18 columns

Check the data types of the columns for the dataset.

data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 56926 entries, 0 to 56925
Data columns (total 18 columns):
 #   Column                                Non-Null Count  Dtype  
---  ------                                --------------  -----  
 0   no_of_adults                          56926 non-null  int64  
 1   no_of_children                        56926 non-null  int64  
 2   no_of_weekend_nights                  56926 non-null  int64  
 3   no_of_week_nights                     56926 non-null  int64  
 4   type_of_meal_plan                     56926 non-null  object 
 5   required_car_parking_space            56926 non-null  int64  
 6   room_type_reserved                    56926 non-null  object 
 7   lead_time                             56926 non-null  int64  
 8   arrival_year                          56926 non-null  int64  
 9   arrival_month                         56926 non-null  int64  
 10  arrival_date                          56926 non-null  int64  
 11  market_segment_type                   56926 non-null  object 
 12  repeated_guest                        56926 non-null  int64  
 13  no_of_previous_cancellations          56926 non-null  int64  
 14  no_of_previous_bookings_not_canceled  56926 non-null  int64  
 15  avg_price_per_room                    56926 non-null  float64
 16  no_of_special_requests                56926 non-null  int64  
 17  booking_status                        56926 non-null  object 
dtypes: float64(1), int64(13), object(4)
memory usage: 7.8+ MB

Observations:

• Most of the data-types are int64 and one of the columns has data-type float64.
• 4 columns - type_of_meal_plan, room_type_reserved, market_segment_type and booking_status are having data-types as an object, this means we need to convert these into suitable data-type before we feed our data into the model.

Data Preprocessing

Let's check the duplicate data. And if any, we should remove it.

data[data.duplicated()].count()

no_of_adults                            14350
no_of_children                          14350
no_of_weekend_nights                    14350
no_of_week_nights                       14350
type_of_meal_plan                       14350
required_car_parking_space              14350
room_type_reserved                      14350
lead_time                               14350
arrival_year                            14350
arrival_month                           14350
arrival_date                            14350
market_segment_type                     14350
repeated_guest                          14350
no_of_previous_cancellations            14350
no_of_previous_bookings_not_canceled    14350
avg_price_per_room                      14350
no_of_special_requests                  14350
booking_status                          14350
dtype: int64

• There are 14350 duplicate values in the dataset

Let's drop the duplicate values

data.drop_duplicates(inplace=True)

Check for missing values

data.isnull().sum()

no_of_adults                            0
no_of_children                          0
no_of_weekend_nights                    0
no_of_week_nights                       0
type_of_meal_plan                       0
required_car_parking_space              0
room_type_reserved                      0
lead_time                               0
arrival_year                            0
arrival_month                           0
arrival_date                            0
market_segment_type                     0
repeated_guest                          0
no_of_previous_cancellations            0
no_of_previous_bookings_not_canceled    0
avg_price_per_room                      0
no_of_special_requests                  0
booking_status                          0
dtype: int64

• There are no missing values in the data.

Fixing the datatypes

# categorical column should be converted to categorical type
# (It reduces the data space required to store the dataframe,
# every class in the categorical column will be represented by a number under the hood.
# This is useful during model building)

data["type_of_meal_plan"] = data.type_of_meal_plan.astype("category")
data["room_type_reserved"] = data.room_type_reserved.astype("category")
data["market_segment_type"] = data.market_segment_type.astype("category")
data["booking_status"] = data.booking_status.astype("category")

data.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 42576 entries, 0 to 56924
Data columns (total 18 columns):
 #   Column                                Non-Null Count  Dtype   
---  ------                                --------------  -----   
 0   no_of_adults                          42576 non-null  int64   
 1   no_of_children                        42576 non-null  int64   
 2   no_of_weekend_nights                  42576 non-null  int64   
 3   no_of_week_nights                     42576 non-null  int64   
 4   type_of_meal_plan                     42576 non-null  category
 5   required_car_parking_space            42576 non-null  int64   
 6   room_type_reserved                    42576 non-null  category
 7   lead_time                             42576 non-null  int64   
 8   arrival_year                          42576 non-null  int64   
 9   arrival_month                         42576 non-null  int64   
 10  arrival_date                          42576 non-null  int64   
 11  market_segment_type                   42576 non-null  category
 12  repeated_guest                        42576 non-null  int64   
 13  no_of_previous_cancellations          42576 non-null  int64   
 14  no_of_previous_bookings_not_canceled  42576 non-null  int64   
 15  avg_price_per_room                    42576 non-null  float64 
 16  no_of_special_requests                42576 non-null  int64   
 17  booking_status                        42576 non-null  category
dtypes: category(4), float64(1), int64(13)
memory usage: 5.0 MB

Summary of the dataset

data.describe().T

	count	mean	std	min	25%	50%	75%	max
no_of_adults	42576.0	1.916737	0.527524	0.0	2.0	2.0	2.0	4.0
no_of_children	42576.0	0.142146	0.459920	0.0	0.0	0.0	0.0	10.0
no_of_weekend_nights	42576.0	0.895270	0.887864	0.0	0.0	1.0	2.0	8.0
no_of_week_nights	42576.0	2.321167	1.519328	0.0	1.0	2.0	3.0	17.0
required_car_parking_space	42576.0	0.034362	0.182160	0.0	0.0	0.0	0.0	1.0
lead_time	42576.0	77.315953	77.279616	0.0	16.0	53.0	118.0	521.0
arrival_year	42576.0	2018.297891	0.626126	2017.0	2018.0	2018.0	2019.0	2019.0
arrival_month	42576.0	6.365488	3.051924	1.0	4.0	6.0	9.0	12.0
arrival_date	42576.0	15.682873	8.813991	1.0	8.0	16.0	23.0	31.0
repeated_guest	42576.0	0.030886	0.173011	0.0	0.0	0.0	0.0	1.0
no_of_previous_cancellations	42576.0	0.025413	0.358194	0.0	0.0	0.0	0.0	13.0
no_of_previous_bookings_not_canceled	42576.0	0.222731	2.242308	0.0	0.0	0.0	0.0	72.0
avg_price_per_room	42576.0	112.375800	40.865896	0.0	85.5	107.0	135.0	540.0
no_of_special_requests	42576.0	0.768109	0.837264	0.0	0.0	1.0	1.0	5.0

Observations:

• no_of_adults have the Mean and medium values as 1.91 and 2 which are very close values. The min no_of_adults is 0 and max value is 4.
• no_of_weekend_nights, no_of_week_nights and no_of_special_requests have outliers towards the right. They have very close Mean and medium values.
• For lead_time the mean 77.31 is higher than the median 53 and the data is skewed to the right and have outliers.
• arrival_year, arrival_month and arrival_date have an almost symmetrical distribution with almost equal mean and median values.
• repeated_guest and required_car_parking_space have most of their values as 0. There are some outliers with value as 1.
• no_of_children, no_of_previous_cancellations and no_of_previous_bookings_not_canceled have most of their values as 0. They have ouliers to the right.
• For avg_price_per_room the mean is 112.37 and median is 107. The data is skewed towards the right and there are outliers.

data.describe(include=["category"])

	type_of_meal_plan	room_type_reserved	market_segment_type	booking_status
count	42576	42576	42576	42576
unique	4	7	5	2
top	Meal Plan 1	Room_Type 1	Online	Not_Canceled
freq	31863	29730	34169	28089

cat_columns = [
    "type_of_meal_plan",
    "room_type_reserved",
    "market_segment_type",
    "booking_status",
]
for i in cat_columns:
    print(data[i].value_counts())
    print("*" * 50)

Meal Plan 1     31863
Not Selected     8716
Meal Plan 2      1989
Meal Plan 3         8
Name: type_of_meal_plan, dtype: int64
**************************************************
Room_Type 1    29730
Room_Type 4     9369
Room_Type 6     1540
Room_Type 5      906
Room_Type 2      718
Room_Type 7      307
Room_Type 3        6
Name: room_type_reserved, dtype: int64
**************************************************
Online           34169
Offline           5777
Corporate         1939
Complementary      496
Aviation           195
Name: market_segment_type, dtype: int64
**************************************************
Not_Canceled    28089
Canceled        14487
Name: booking_status, dtype: int64
**************************************************

Observations:

• Meal Plan 1 is the most preferred meal selected in 31863 bookings and Meal Plan 3 is the least preferred meal only used in 8 bookings.
• Room_Type 1 is the room mostly selceted by 29730 bookings whereas Room_Type 3 is only selected in 6 bookings.
• Online is the mostly used market_segment type with 34169 bookings. The least used is Aviation with only 195 bookings.
• 28089 Bookings were not cancelled whereas 14487 bookings were cancelled.

Exploratory Data Analysis (EDA)

Univariate Analysis

Let us explore the numerical variables first

# function to plot a boxplot and a histogram along the same scale.


def histogram_boxplot(data, feature, figsize=(12, 7), kde=False, bins=None):
    """
    Boxplot and histogram combined

    data: dataframe
    feature: dataframe column
    figsize: size of figure (default (12,7))
    kde: whether to the show density curve (default False)
    bins: number of bins for histogram (default None)
    """
    f2, (ax_box2, ax_hist2) = plt.subplots(
        nrows=2,  # Number of rows of the subplot grid= 2
        sharex=True,  # x-axis will be shared among all subplots
        gridspec_kw={"height_ratios": (0.25, 0.75)},
        figsize=figsize,
    )  # creating the 2 subplots
    sns.boxplot(
        data=data, x=feature, ax=ax_box2, showmeans=True, color="violet"
    )  # boxplot will be created and a star will indicate the mean value of the column
    sns.histplot(
        data=data, x=feature, kde=kde, ax=ax_hist2, bins=bins, palette="winter"
    ) if bins else sns.histplot(
        data=data, x=feature, kde=kde, ax=ax_hist2
    )  # For histogram
    ax_hist2.axvline(
        data[feature].mean(), color="green", linestyle="--"
    )  # Add mean to the histogram
    ax_hist2.axvline(
        data[feature].median(), color="black", linestyle="-"
    )  # Add median to the histogram

Univariate Analysis of lead_time

histogram_boxplot(data, "lead_time")

Observations

• Mean is around 77 and greater than median which is 53.
• The graph is skewed towards the right and there are many outliers towards the right.
• Majority of the bookings have have lead_time of 0 days

Univariate Analysis of no_of_previous_bookings_not_canceled

histogram_boxplot(data, "no_of_previous_bookings_not_canceled")

Observations

• Majority of the previous bookings for customers have not been cancelled.
• But there are many outliers towards the right.

Univariate Analysis of avg_price_per_room

histogram_boxplot(data, "avg_price_per_room")

Observations

• The mean avg_price_per_room is 112.37 and median is 107.
• There are few outliers to the left and many to the right.
• There are many more than 500 rooms with 0 price.
• Most of the rooms are prices between 50 and 200

# function to create labeled barplots


def labeled_barplot(data, feature, perc=False, n=None):
    """
    Barplot with percentage at the top

    data: dataframe
    feature: dataframe column
    perc: whether to display percentages instead of count (default is False)
    n: displays the top n category levels (default is None, i.e., display all levels)
    """

    total = len(data[feature])  # length of the column
    count = data[feature].nunique()
    if n is None:
        plt.figure(figsize=(count + 2, 6))
    else:
        plt.figure(figsize=(n + 2, 6))

    plt.xticks(rotation=90, fontsize=15)
    ax = sns.countplot(
        data=data,
        x=feature,
        palette="Paired",
        order=data[feature].value_counts().index[:n].sort_values(),
    )

    for p in ax.patches:
        if perc == True:
            label = "{:.1f}%".format(
                100 * p.get_height() / total
            )  # percentage of each class of the category
        else:
            label = p.get_height()  # count of each level of the category

        x = p.get_x() + p.get_width() / 2  # width of the plot
        y = p.get_height()  # height of the plot

        ax.annotate(
            label,
            (x, y),
            ha="center",
            va="center",
            size=12,
            xytext=(0, 5),
            textcoords="offset points",
        )  # annotate the percentage

    plt.show()  # show the plot

Univariate Analysis of no_of_adults

labeled_barplot(data, "no_of_adults", perc=True)

Observations

• 73% of the bookings have no_of_adults as 2.
• 0.4% have no_of_adults as 0. We will check further if these entries needs to be removed or not.
• 0.1% bookings have no_of_adults as 4.

Univariate Analysis of no_of_children

labeled_barplot(data, "no_of_children", perc=True)

Observations

• 90% bookings have 0 no_of_children.
• 0.1% or less bookings have 3,9 or 10 children.

Univariate Analysis of no_of_weekend_nights

labeled_barplot(data, "no_of_weekend_nights", perc=True)

Observations

• 41.9% bookings have 0 no_of_weekend_nights.
• 0.1% or less bookings have 5,6,7 or 8 no_of_weekend_nights.

Univariate Analysis of no_of_week_nights

labeled_barplot(data, "no_of_week_nights", perc=True)

Observations

• The highest no_of_week_nights booked is 2 with 27.6% bookings.
• The least no_of_week_nights booked are 7 to 17 days with 0.4% bookings or less.

Univariate Analysis of required_car_parking_space

labeled_barplot(data, "required_car_parking_space", perc=True)

Observations

• 96.6% bookings do not required_car_parking_space.
• Only 3.4% bookings required_car_parking_space.

Univariate Analysis of arrival_year

labeled_barplot(data, "arrival_year", perc=True)

Observations

• 51.9% bookings have guests arriving in 2018.
• Only 9.1% will arrive in 2017.

Univariate Analysis of arrival_month

labeled_barplot(data, "arrival_month", perc=True)

Observations

• Most of the bookings 12.5% are made for the month of August.
• Least number of bookings 5% are made for January.

Univariate Analysis of arrival_date

labeled_barplot(data, "arrival_date", perc=True)

Observations

• Arrival_data is almost evenly distributed acroos all dates and do not seem to be significant.

Univariate Analysis of repeated_guest

labeled_barplot(data, "repeated_guest", perc=True)

Observations

• 96.9% bookings are not made by repeated_guest.
• Only 3.1% bookings are made by repeated_guest.

Univariate Analysis of no_of_previous_cancellations

labeled_barplot(data, "no_of_previous_cancellations", perc=True)

Observations

• 99% bookings have no previous cancellations.
• 0.1% and less bookings have 3,4,5,6,11 and even 13 previous cancellations.

Univariate Analysis of no_of_special_requests

labeled_barplot(data, "no_of_special_requests", perc=True)

Observations

• 45.2% bookings have no special requests.
• Less than 0.1% bookings have made 5 special requests which is the highest.

Let us now explore the categorical variables

Univariate Analysis of type_of_meal_plan

labeled_barplot(data, "type_of_meal_plan", perc=True)

Observations

• Meal Plan 1 is the most preferred meal chosen by 74.8% bookings.
• Meal Plan 3 is the least preferred meal chosen less than 0.1% bookings.
• 20.5% bookings have not selected any meal plan.

Univariate Analysis of room_type_reserved

labeled_barplot(data, "room_type_reserved", perc=True)

Observations

• Room_type 1 is the most preferred room chosen by 69.8% bookings.
• Room_type 3 is the least preferred room chosen less than 0.1% bookings.

Univariate Analysis of market_segment_type

labeled_barplot(data, "market_segment_type", perc=True)

Observations

• 80.3% bookings are made by Online Market_segment_type.
• The least 0.5% bookings are made by Aviation Market_segment_type.

Univariate Analysis of booking_status

labeled_barplot(data, "booking_status", perc=True)

Observations

• 66% of the bookings are not_cancelled.
• 34% of the bookings are cancelled.

Bivariate Analysis

Plot bivariate charts between numeric variables to understand their interaction with each other.

Correlation

plt.figure(figsize=(15, 7))
sns.heatmap(data.corr(), annot=True, vmin=-1, vmax=1, cmap="Spectral")
plt.show()

Observations

• There are no significant positive or negative correlation between any variables.
• no_of_previous_bookings_not_canceled has a strong positive correlation with repeated_guest and no_of_previous_cancellations.

Bivariate Scatter Plots

sns.pairplot(data=data, diag_kind="kde")
plt.show()

Observations

• We do not see any significant relation between any variables through scatter plots.

Relationship of Room Price and Market Segment

plt.figure(figsize=(10, 5))
sns.barplot(data=data, x="market_segment_type", y="avg_price_per_room")

<AxesSubplot:xlabel='market_segment_type', ylabel='avg_price_per_room'>

Observations

• Customers with Online Market Segment Type have the highest average price per room which is around 120 euros.
• Customers with Complementary Offers have to pay as low as 0 euros per room.

Relationship of repeated_guest and booking_status

# crosstab

ax = pd.crosstab(data["repeated_guest"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("repeated_guest")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
repeated_guest
0	35.086401	64.913599
1	0.760456	99.239544

Observations

• Only 0.76% of repeated guests have cancelled their bookings.

Relationship of no_of_special_requests and booking_status

# crosstab

ax = pd.crosstab(data["no_of_special_requests"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("no_of_special_requests")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
no_of_special_requests
0	45.516954	54.483046
1	27.910860	72.089140
2	21.767748	78.232252
3	0.000000	100.000000
4	0.000000	100.000000
5	0.000000	100.000000

Observations

• For 0 special requests there is only a small difference between cancelled and non-cancelled bookings.
• However as no. of special requests increases, there are big gaps of difffernce between cancelled and non-cancelled bookings. And more customers do not cancel their bookings
• We can conclude that if a customer makes special requests, he is more interested in keeping the booking and less chances of cancelling it.

Relationship of no_of_adults and booking_status

# crosstab

ax = pd.crosstab(data["no_of_adults"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("no_of_adults")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
no_of_adults
0	41.304348	58.695652
1	21.875000	78.125000
2	35.398629	64.601371
3	44.976433	55.023567
4	39.285714	60.714286

Observations

• There do not seem to be any significant relationship between the no.of adults and the booking_status.

Relationship of no_of_children and booking_status

# crosstab

ax = pd.crosstab(data["no_of_children"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("no_of_children")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
no_of_children
0	32.845953	67.154047
1	39.398672	60.601328
2	52.779438	47.220562
3	35.897436	64.102564
9	50.000000	50.000000
10	0.000000	100.000000

Observations

• There do not seem to be any significant relationship between the no.of children and the booking_status.

Relationship of no_of_weekend_nights and booking_status

# crosstab

ax = pd.crosstab(data["no_of_weekend_nights"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("no_of_weekend_nights")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
no_of_weekend_nights
0	31.577767	68.422233
1	33.686786	66.313214
2	36.848252	63.151748
3	53.181818	46.818182
4	68.518519	31.481481
5	70.000000	30.000000
6	67.741935	32.258065
7	100.000000	0.000000
8	100.000000	0.000000

Observations

• As the no_of_weekend_nights increases, the "cancelled" booking_status also have increased.
• Bookings with 7 and 8 weekend nights were all cancelled.

Relationship of no_of_week_nights and booking_status

# crosstab

ax = pd.crosstab(data["no_of_week_nights"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("no_of_week_nights")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
no_of_week_nights
0	24.705041	75.294959
1	27.856226	72.143774
2	33.823529	66.176471
3	36.055901	63.944099
4	41.199226	58.800774
5	44.071856	55.928144
6	53.488372	46.511628
7	54.545455	45.454545
8	61.157025	38.842975
9	60.416667	39.583333
10	84.042553	15.957447
11	85.000000	15.000000
12	68.750000	31.250000
13	77.777778	22.222222
14	50.000000	50.000000
15	57.142857	42.857143
16	71.428571	28.571429
17	66.666667	33.333333

Observations

• As the no_of_week_nights increases, the "cancelled" booking_status also have increased steadily till 11 nights to 85% cancellations.
• After that we see some exception for 14 and 15 where cancelled status has decreased to 57%
• But overall we can see from graph that no_of_week_nights has a strong positive correlation with cancelled status.

Relationship of required_car_parking_space and booking_status

# crosstab

ax = pd.crosstab(data["required_car_parking_space"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("required_car_parking_space")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
required_car_parking_space
0	34.799212	65.200788
1	12.303486	87.696514

Observations

• 35% of bookings that do not require car parking space are cancelled.
• 12% of bookings that require a car parking space are cancelled.

Relationship of lead_time and booking_status

plt.figure(figsize=(10, 5))
sns.boxplot(y="lead_time", x="booking_status", data=data)
plt.show()

Observations

• Cancelled bookings have 25% to 75% lead time between 50 and 180 days. There are many outliers after 400.
• Not cancelled bookings have 25% to 75% of the lead time between 20 to 100 days. There are many outliers after 200.

Relationship of arrival_year and booking_status

# crosstab

ax = pd.crosstab(data["arrival_year"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("arrival_year")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
arrival_year
2017	12.227074	87.772926
2018	31.510381	68.489619
2019	42.501207	57.498793

Observations

• As the arrival year has increased, the cancellation percentage has also increased.

Relationship of arrival_month and booking_status

# crosstab

ax = pd.crosstab(data["arrival_month"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(10, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("arrival_month")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
arrival_month
1	12.009456	87.990544
2	27.552786	72.447214
3	29.549951	70.450049
4	38.490655	61.509345
5	38.500460	61.499540
6	38.890253	61.109747
7	47.407407	52.592593
8	46.592620	53.407380
9	29.048086	70.951914
10	28.607043	71.392957
11	22.627737	77.372263
12	14.255765	85.744235

Observations

• The cancelled status increased steadily from January to August.
• From September to December there is decline in the cancelled status.
• July and August are the months with most number of cancellations in terms of percentage.
• In short we can say summer months are seeing more cancellations.

Relationship of arrival_data and booking_status

# crosstab

ax = pd.crosstab(data["arrival_date"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(20, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("arrival_date")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
arrival_date
1	34.624146	65.375854
2	31.725050	68.274950
3	34.985836	65.014164
4	33.476703	66.523297
5	30.565248	69.434752
6	34.287759	65.712241
7	34.773371	65.226629
8	35.327635	64.672365
9	32.666190	67.333810
10	33.746356	66.253644
11	32.244898	67.755102
12	34.219734	65.780266
13	32.466620	67.533380
14	33.040702	66.959298
15	35.940803	64.059197
16	35.388928	64.611072
17	36.485580	63.514420
18	33.187773	66.812227
19	29.298487	70.701513
20	32.081911	67.918089
21	33.738385	66.261615
22	36.687797	63.312203
23	33.775351	66.224649
24	36.980831	63.019169
25	37.269939	62.730061
26	36.078965	63.921035
27	34.583902	65.416098
28	34.475375	65.524625
29	34.767523	65.232477
30	34.730056	65.269944
31	30.875000	69.125000

Observations

• We do not see any pattern in arrival date and booking_status.

Relationship of no_of_previous_cancellations and booking_status

# crosstab
ax = pd.crosstab(data["no_of_previous_cancellations"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(20, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("no_of_previous_cancellations")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
no_of_previous_cancellations
0	34.361056	65.638944
1	3.212851	96.787149
2	0.000000	100.000000
3	2.127660	97.872340
4	0.000000	100.000000
5	0.000000	100.000000
6	0.000000	100.000000
11	0.000000	100.000000
13	100.000000	0.000000

Observations

• There do not seem to be any significant relation between no_of_previous_cancellations and booking_status

Relationship of no_of_previous_bookings_not_canceled and booking_status

plt.figure(figsize=(40, 10))
sns.countplot(x="no_of_previous_bookings_not_canceled", hue="booking_status", data=data)
plt.show()

Observations

• As the no_of_previous_bookings_not_canceled increases, most bookings are in not_cancelled status.

Relationship of avg_price_per_room and booking_status

plt.figure(figsize=(10, 5))
sns.boxplot(y="avg_price_per_room", x="booking_status", data=data)
plt.show()

Observations

• 25% to 75% of the avg_price_per_room for cancelled bookings are in the range of 100 to 150 euros.
• 25% to 75% of the avg_price_per_room for not_cancelled bookings are in the range of 90 to 130 euros.
• There are outliers for both booking_status but more outliers are for the cancelled bookings.

Relationship of type_of_meal_plan and booking_status

# crosstab
ax = pd.crosstab(data["type_of_meal_plan"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(20, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("type_of_meal_plan")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
type_of_meal_plan
Meal Plan 1	32.988105	67.011895
Meal Plan 2	43.086978	56.913022
Meal Plan 3	12.500000	87.500000
Not Selected	35.773291	64.226709

Observations

• Most customers whcih is 43% who have taken Meal Plan 2 have cancelled their bookings.
• For Meal Plan 3 only 12% have cancelled their bookings.

Relationship of room_type_reserved and booking_status

# crosstab
ax = pd.crosstab(data["room_type_reserved"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(20, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("room_type_reserved")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
room_type_reserved
Room_Type 1	31.029263	68.970737
Room_Type 2	38.161560	61.838440
Room_Type 3	33.333333	66.666667
Room_Type 4	39.310492	60.689508
Room_Type 5	40.507726	59.492274
Room_Type 6	53.636364	46.363636
Room_Type 7	35.830619	64.169381

Observations

• Most customers which is 54% who chose Room Type 6 have cancelled their bookings.
• Room Type 1 only 31% customers have cancelled their bookings.

Relationship of market_segment_type and booking_status

# crosstab
ax = pd.crosstab(data["market_segment_type"], data["booking_status"]).apply(
    lambda r: r / r.sum() * 100, axis=1
)
ax_1 = ax.plot.bar(figsize=(20, 5), stacked=True, rot=0)
display(ax)

plt.legend(loc="upper center", bbox_to_anchor=(0.1, 1.0), title="booking_status")

plt.xlabel("market_segment_type")
plt.ylabel("Percent Distribution")

for rec in ax_1.patches:
    height = rec.get_height()
    ax_1.text(
        rec.get_x() + rec.get_width() / 2,
        rec.get_y() + height / 2,
        "{:.0f}%".format(height),
        ha="center",
        va="bottom",
    )

plt.show()

booking_status	Canceled	Not_Canceled
market_segment_type
Aviation	16.923077	83.076923
Complementary	0.000000	100.000000
Corporate	8.612687	91.387313
Offline	13.917258	86.082742
Online	39.459744	60.540256

Observations

• Customers who belong to market segment online have the most high percentage to cancel which is 39%.
• Customers who belong to market segment Complimentary have the least high percentage to cancel which is 0%.

Insights Based on EDA

• Customers with Online Market Segment Type have the highest average price per room which is around 120 euros.
• Only 0.76% of repeated guests have cancelled their bookings. So a Repeated guests have higher chances of not Cancelling their booking.
• As no. of special requests increases, there are big gaps of difffernce between cancelled and non-cancelled bookings. And more customers do not cancel their bookings. We can conclude that if a customer makes special requests, he is more interested in keeping the booking and less chances of cancelling it.
• No_of_Adults and no_of_children do not have any significance on the booking_status
• As the no_of_weekend_nights increases, the "cancelled" booking_status also have increased. Bookings with 7 and 8 weekend nights were all cancelled.
• As the no_of_week_nights increases, the "cancelled" booking_status also have increased steadily till 11 nights to 85% cancellations. Overall we can see from graph that no_of_week_nights has a strong positive correlation with cancelled status.
• 35% of bookings that do not require car parking space are cancelled. Only 12% of bookings that require a car parking space are cancelled.
• The higher the lead time and arrival year, higher the chances of cancellation.
• The cancelled status increased steadily from January to August.From September to December there is decline in the cancelled status. July and August are the months with most number of cancellations in terms of percentage. In short we can say summer months are seeing more cancellations.
• If the previous bookings were not cancelled , there are high chances that current booking will not be cancelled.
• Cancelled bookings contains more rooms with high prices. So high room price is a factor for cancellation.
• Most customers whcih is 43% who have taken Meal Plan 2 have cancelled their bookings. For Meal Plan 3 only 12% have cancelled their bookings.
• Most customers which is 54% who chose Room Type 6 have cancelled their bookings. Room Type 1 only 31% customers have cancelled their bookings.
• Customers who belong to market segment online have the most high percentage to cancel which is 39%. Customers who belong to market segment Complimentary have the least high percentage to cancel which is 0%.

Data Preprocessing (contd.)

Lets see if there are any rows with both no_of_weekend_nights and no_of_week_nights as 0 and remove them

len(data[(data["no_of_weekend_nights"] == 0) & (data["no_of_week_nights"] == 0)])

99

# drop these rows
data.drop(data[(data["no_of_weekend_nights"] == 0) & (data["no_of_week_nights"] == 0)].index,
    axis=0,
    inplace=True,
)

data.shape

(42477, 18)

Lets see if there are any rows with both no_of_adults and no_of_children as 0

len(data[(data["no_of_adults"] == 0) & (data["no_of_children"] == 0)])

0

Lets see if there are any rows with no_of_adults 0 and non 0 no_of_children

len(data[(data["no_of_adults"] == 0) & (data["no_of_children"] != 0)])

184

Since we do not know the age of the children and if these bookings are made under special conditions , we will not remove these rows.

Outlier Detection

• We have only two numerical variable that are continous in nature : lead_time and avg_price_per_room
• Lets Detect their Outliers and Remove them

# let's plot the boxplots of these columns to check for outliers
plt.figure(figsize=(17, 1))
sns.boxplot(data=data, x="lead_time")

<AxesSubplot:xlabel='lead_time'>

# let's plot the boxplots of these columns to check for outliers
plt.figure(figsize=(17, 1))
sns.boxplot(data=data, x="avg_price_per_room")

<AxesSubplot:xlabel='avg_price_per_room'>

Outlier Treatment

def treat_outliers(df, col):
    """
    treats outliers in a variable
    col: str, name of the numerical variable
    df: dataframe
    col: name of the column
    """
    Q1 = df[col].quantile(0.25)  # 25th quantile
    Q3 = df[col].quantile(0.75)  # 75th quantile
    IQR = Q3 - Q1
    Lower_Whisker = Q1 - 1.5 * IQR
    Upper_Whisker = Q3 + 1.5 * IQR

    # all the values smaller than Lower_Whisker will be assigned the value of Lower_Whisker
    # all the values greater than Upper_Whisker will be assigned the value of Upper_Whisker
    df[col] = np.clip(df[col], Lower_Whisker, Upper_Whisker)

    return df


def treat_outliers_all(df, col_list):
    """
    treat outlier in all numerical variables
    col_list: list of numerical variables
    df: data frame
    """
    for c in col_list:
        df = treat_outliers(df, c)

    return df

# treating the outliers
numerical_col = ["avg_price_per_room", "lead_time"]
data = treat_outliers_all(data, numerical_col)

# let's plot the boxplots of these columns to check for outliers
plt.figure(figsize=(17, 1))
sns.boxplot(data=data, x="lead_time")

<AxesSubplot:xlabel='lead_time'>

# let's plot the boxplots of these columns to check for outliers
plt.figure(figsize=(17, 1))
sns.boxplot(data=data, x="avg_price_per_room")

<AxesSubplot:xlabel='avg_price_per_room'>

Building a Logistic Regression model

Data Preparation

Lets convert the booking_status variable to a numerical variable.

• Replace Cancelled to 1
• Replace Not Cancelled to 0

data["booking_status"] = data["booking_status"].apply(
    lambda x: 1 if x == "Canceled" else 0
)

Before we proceed to build a model, we'll have to encode categorical features.

# creating dummy varibles
dummy_data = pd.get_dummies(
    data,
    columns=["type_of_meal_plan", "room_type_reserved", "market_segment_type",],
    drop_first=True,
)
dummy_data.head()

	no_of_adults	no_of_children	no_of_weekend_nights	no_of_week_nights	required_car_parking_space	lead_time	arrival_year	arrival_month	arrival_date	repeated_guest	no_of_previous_cancellations	no_of_previous_bookings_not_canceled	avg_price_per_room	no_of_special_requests	booking_status	type_of_meal_plan_Meal Plan 2	type_of_meal_plan_Meal Plan 3	type_of_meal_plan_Not Selected	room_type_reserved_Room_Type 2	room_type_reserved_Room_Type 3	room_type_reserved_Room_Type 4	room_type_reserved_Room_Type 5	room_type_reserved_Room_Type 6	room_type_reserved_Room_Type 7	market_segment_type_Complementary	market_segment_type_Corporate	market_segment_type_Offline	market_segment_type_Online
0	2	0	1	2	0	224	2017	10	2	0	0	0	65.00	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0
1	2	0	2	3	0	5	2018	11	6	0	0	0	106.68	1	0	0	0	1	0	0	0	0	0	0	0	0	0	1
2	1	0	2	1	0	1	2018	2	28	0	0	0	60.00	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1
3	2	0	0	2	0	211	2018	5	20	0	0	0	100.00	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1
4	3	0	0	3	0	271	2019	7	13	0	0	0	89.10	2	1	0	0	1	0	0	0	0	0	0	0	0	0	1

Split Data

We'll split the data into train and test to be able to evaluate the model that we build on the train data.

X = dummy_data.drop("booking_status", axis=1)  # Features
y = dummy_data["booking_status"]  # Labels (Target Variable)

# Splitting data into training and test set:
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1)
print(X_train.shape, X_test.shape)

(29733, 27) (12744, 27)

print("Number of rows in train data =", X_train.shape[0])
print("Number of rows in test data =", X_test.shape[0])

Number of rows in train data = 29733
Number of rows in test data = 12744

print("Percentage of classes in training set:")
print(y_train.value_counts(normalize=True))
print("Percentage of classes in test set:")
print(y_test.value_counts(normalize=True))

Percentage of classes in training set:
0    0.661588
1    0.338412
Name: booking_status, dtype: float64
Percentage of classes in test set:
0    0.652935
1    0.347065
Name: booking_status, dtype: float64

Model evaluation criterion

Model can make wrong predictions as:

1. Predicting a customer will not cancel the booking but in reality the customer cancels the booking. - Loss of revenue

2. Predicting a customer will cancel the booking but in reality the customer did not cancel the booking. - spend more money for marketing and reduce prices.

Which case is more important?

• Predicting a customer will not cancel the booking but in reality the customer cancels the booking. - Loss of revenue

How to reduce this loss i.e need to reduce False Negatives?

• recall should be maximized, the greater the recall higher the chances of minimizing the false negatives.

First, let's create functions to calculate different metrics and confusion matrix so that we don't have to use the same code repeatedly for each model.

• The model_performance_classification_statsmodels function will be used to check the model performance of models.
• The confusion_matrix_statsmodels function will be used to plot confusion matrix.

# defining a function to compute different metrics to check performance of a classification model built using statsmodels
def model_performance_classification_statsmodels(
    model, predictors, target, threshold=0.5
):
    """
    Function to compute different metrics to check classification model performance

    model: classifier
    predictors: independent variables
    target: dependent variable
    threshold: threshold for classifying the observation as class 1
    """

    # checking which probabilities are greater than threshold
    pred_temp = model.predict(predictors) > threshold
    # rounding off the above values to get classes
    pred = np.round(pred_temp)

    acc = accuracy_score(target, pred)  # to compute Accuracy
    recall = recall_score(target, pred)  # to compute Recall
    precision = precision_score(target, pred)  # to compute Precision
    f1 = f1_score(target, pred)  # to compute F1-score

    # creating a dataframe of metrics
    df_perf = pd.DataFrame(
        {"Accuracy": acc, "Recall": recall, "Precision": precision, "F1": f1,},
        index=[0],
    )

    return df_perf

# defining a function to plot the confusion_matrix of a classification model


def confusion_matrix_statsmodels(model, predictors, target, threshold=0.5):
    """
    To plot the confusion_matrix with percentages

    model: classifier
    predictors: independent variables
    target: dependent variable
    threshold: threshold for classifying the observation as class 1
    """
    y_pred = model.predict(predictors) > threshold
    cm = confusion_matrix(target, y_pred)
    labels = np.asarray(
        [
            ["{0:0.0f}".format(item) + "\n{0:.2%}".format(item / cm.flatten().sum())]
            for item in cm.flatten()
        ]
    ).reshape(2, 2)

    plt.figure(figsize=(6, 4))
    sns.heatmap(cm, annot=labels, fmt="")
    plt.ylabel("True label")
    plt.xlabel("Predicted label")

##  Function to calculate recall score
def get_recall_score(model, predictors, target):
    """
    model: classifier
    predictors: independent variables
    target: dependent variable
    
    """
    prediction = model.predict(predictors)
    return recall_score(target, prediction)

Checking Multicollinearity

• In order to make statistical inferences from a logistic regression model, it is important to ensure that there is no multicollinearity present in the data.
• If VIF is between 1 and 5, then there is low multicollinearity.
• If VIF is between 5 and 10, we say there is moderate multicollinearity.
• If VIF is exceeding 10, it shows signs of high multicollinearity.

from statsmodels.stats.outliers_influence import variance_inflation_factor

# we will define a function to check VIF
def checking_vif(predictors):
    vif = pd.DataFrame()
    vif["feature"] = predictors.columns

    # calculating VIF for each feature
    vif["VIF"] = [
        variance_inflation_factor(predictors.values, i)
        for i in range(len(predictors.columns))
    ]
    return vif

checking_vif(X_train)

	feature	VIF
0	no_of_adults	20.409879
1	no_of_children	2.348806
2	no_of_weekend_nights	2.192104
3	no_of_week_nights	3.779718
4	required_car_parking_space	1.072203
5	lead_time	2.433912
6	arrival_year	261.122669
7	arrival_month	5.574838
8	arrival_date	4.175151
9	repeated_guest	2.108817
10	no_of_previous_cancellations	1.602796
11	no_of_previous_bookings_not_canceled	1.995322
12	avg_price_per_room	21.323715
13	no_of_special_requests	2.046411
14	type_of_meal_plan_Meal Plan 2	1.140941
15	type_of_meal_plan_Meal Plan 3	1.024754
16	type_of_meal_plan_Not Selected	1.587943
17	room_type_reserved_Room_Type 2	1.119320
18	room_type_reserved_Room_Type 3	1.001401
19	room_type_reserved_Room_Type 4	1.851392
20	room_type_reserved_Room_Type 5	1.144932
21	room_type_reserved_Room_Type 6	2.181760
22	room_type_reserved_Room_Type 7	1.152450
23	market_segment_type_Complementary	3.757636
24	market_segment_type_Corporate	12.024021
25	market_segment_type_Offline	34.638314
26	market_segment_type_Online	195.388423

• Remove the Variable with High VIF.
• Repeat the Steps until we get no variables with VIF>5.

X_train1 = X_train.drop(["arrival_year",], axis=1,)

checking_vif(X_train1)

	feature	VIF
0	no_of_adults	20.311008
1	no_of_children	2.348672
2	no_of_weekend_nights	2.184122
3	no_of_week_nights	3.743997
4	required_car_parking_space	1.072202
5	lead_time	2.432830
6	arrival_month	5.518603
7	arrival_date	4.122374
8	repeated_guest	2.064069
9	no_of_previous_cancellations	1.602746
10	no_of_previous_bookings_not_canceled	1.992959
11	avg_price_per_room	20.549886
12	no_of_special_requests	2.044915
13	type_of_meal_plan_Meal Plan 2	1.138549
14	type_of_meal_plan_Meal Plan 3	1.024746
15	type_of_meal_plan_Not Selected	1.582757
16	room_type_reserved_Room_Type 2	1.118535
17	room_type_reserved_Room_Type 3	1.001400
18	room_type_reserved_Room_Type 4	1.849935
19	room_type_reserved_Room_Type 5	1.142597
20	room_type_reserved_Room_Type 6	2.173124
21	room_type_reserved_Room_Type 7	1.149852
22	market_segment_type_Complementary	1.328489
23	market_segment_type_Corporate	2.447333
24	market_segment_type_Offline	5.035240
25	market_segment_type_Online	29.620677

X_train2 = X_train1.drop(["avg_price_per_room",], axis=1,)

checking_vif(X_train2)

	feature	VIF
0	no_of_adults	19.027817
1	no_of_children	2.293127
2	no_of_weekend_nights	2.175730
3	no_of_week_nights	3.728549
4	required_car_parking_space	1.071022
5	lead_time	2.402572
6	arrival_month	5.392907
7	arrival_date	4.119909
8	repeated_guest	2.063750
9	no_of_previous_cancellations	1.602225
10	no_of_previous_bookings_not_canceled	1.992799
11	no_of_special_requests	2.040749
12	type_of_meal_plan_Meal Plan 2	1.086024
13	type_of_meal_plan_Meal Plan 3	1.024696
14	type_of_meal_plan_Not Selected	1.561102
15	room_type_reserved_Room_Type 2	1.114061
16	room_type_reserved_Room_Type 3	1.001362
17	room_type_reserved_Room_Type 4	1.695970
18	room_type_reserved_Room_Type 5	1.070637
19	room_type_reserved_Room_Type 6	2.037480
20	room_type_reserved_Room_Type 7	1.116264
21	market_segment_type_Complementary	1.313848
22	market_segment_type_Corporate	2.238825
23	market_segment_type_Offline	4.515688
24	market_segment_type_Online	22.614131

X_train3 = X_train2.drop(["no_of_adults",], axis=1,)

checking_vif(X_train3)

	feature	VIF
0	no_of_children	2.262371
1	no_of_weekend_nights	2.173338
2	no_of_week_nights	3.726513
3	required_car_parking_space	1.069917
4	lead_time	2.371605
5	arrival_month	5.392444
6	arrival_date	4.119034
7	repeated_guest	2.060639
8	no_of_previous_cancellations	1.602214
9	no_of_previous_bookings_not_canceled	1.992235
10	no_of_special_requests	2.027042
11	type_of_meal_plan_Meal Plan 2	1.085932
12	type_of_meal_plan_Meal Plan 3	1.024691
13	type_of_meal_plan_Not Selected	1.560676
14	room_type_reserved_Room_Type 2	1.106463
15	room_type_reserved_Room_Type 3	1.001362
16	room_type_reserved_Room_Type 4	1.530218
17	room_type_reserved_Room_Type 5	1.051149
18	room_type_reserved_Room_Type 6	2.006950
19	room_type_reserved_Room_Type 7	1.098817
20	market_segment_type_Complementary	1.222798
21	market_segment_type_Corporate	1.952380
22	market_segment_type_Offline	2.608723
23	market_segment_type_Online	11.259872

X_train4 = X_train3.drop(["arrival_month",], axis=1,)

checking_vif(X_train4)

	feature	VIF
0	no_of_children	2.262258
1	no_of_weekend_nights	2.173328
2	no_of_week_nights	3.726225
3	required_car_parking_space	1.069830
4	lead_time	2.340627
5	arrival_date	4.118532
6	repeated_guest	2.060639
7	no_of_previous_cancellations	1.601669
8	no_of_previous_bookings_not_canceled	1.992201
9	no_of_special_requests	2.013531
10	type_of_meal_plan_Meal Plan 2	1.085931
11	type_of_meal_plan_Meal Plan 3	1.024529
12	type_of_meal_plan_Not Selected	1.560381
13	room_type_reserved_Room_Type 2	1.106374
14	room_type_reserved_Room_Type 3	1.001234
15	room_type_reserved_Room_Type 4	1.530201
16	room_type_reserved_Room_Type 5	1.049571
17	room_type_reserved_Room_Type 6	2.006181
18	room_type_reserved_Room_Type 7	1.098807
19	market_segment_type_Complementary	1.173133
20	market_segment_type_Corporate	1.746916
21	market_segment_type_Offline	2.027193
22	market_segment_type_Online	8.523439

X_test4 = X_test[X_train4.columns].astype(float)

• We have removed the variables with VIF>5.
• Since market_segment_type_Online has high VIF but it is part of market_segment_type and its other categoried have low VIF, we are not removing it.
• We will pass the final model Strain4 to Logistic Regression statsmodel

Logistic Regression (with statsmodels library)

logit1 = sm.Logit(y_train, X_train4.astype(float))
lg1 = logit1.fit(
    disp=False
)  # setting disp=False will remove the information on number of iterations

print(lg1.summary())

                           Logit Regression Results                           
==============================================================================
Dep. Variable:         booking_status   No. Observations:                29733
Model:                          Logit   Df Residuals:                    29710
Method:                           MLE   Df Model:                           22
Date:                Fri, 22 Oct 2021   Pseudo R-squ.:                  0.3052
Time:                        20:16:58   Log-Likelihood:                -13220.
converged:                      False   LL-Null:                       -19028.
Covariance Type:            nonrobust   LLR p-value:                     0.000
========================================================================================================
                                           coef    std err          z      P>|z|      [0.025      0.975]
--------------------------------------------------------------------------------------------------------
no_of_children                           0.2458      0.047      5.177      0.000       0.153       0.339
no_of_weekend_nights                    -0.0017      0.018     -0.095      0.924      -0.036       0.033
no_of_week_nights                        0.0648      0.010      6.316      0.000       0.045       0.085
required_car_parking_space              -1.3609      0.114    -11.990      0.000      -1.583      -1.138
lead_time                                0.0161      0.000     64.718      0.000       0.016       0.017
arrival_date                            -0.0050      0.002     -2.902      0.004      -0.008      -0.002
repeated_guest                          -3.9270      0.744     -5.280      0.000      -5.385      -2.469
no_of_previous_cancellations             0.3068      0.105      2.913      0.004       0.100       0.513
no_of_previous_bookings_not_canceled    -0.0419      0.096     -0.438      0.661      -0.229       0.145
no_of_special_requests                  -1.2430      0.023    -54.083      0.000      -1.288      -1.198
type_of_meal_plan_Meal Plan 2            0.3608      0.074      4.905      0.000       0.217       0.505
type_of_meal_plan_Meal Plan 3            8.6767     16.261      0.534      0.594     -23.195      40.548
type_of_meal_plan_Not Selected           0.1615      0.041      3.954      0.000       0.081       0.242
room_type_reserved_Room_Type 2          -0.3746      0.123     -3.041      0.002      -0.616      -0.133
room_type_reserved_Room_Type 3           0.6230      1.561      0.399      0.690      -2.436       3.682
room_type_reserved_Room_Type 4           0.2964      0.039      7.590      0.000       0.220       0.373
room_type_reserved_Room_Type 5           0.6065      0.105      5.784      0.000       0.401       0.812
room_type_reserved_Room_Type 6           0.6191      0.111      5.587      0.000       0.402       0.836
room_type_reserved_Room_Type 7           0.7103      0.188      3.782      0.000       0.342       1.078
market_segment_type_Complementary      -14.6186     19.321     -0.757      0.449     -52.487      23.250
market_segment_type_Corporate           -2.0265      0.114    -17.841      0.000      -2.249      -1.804
market_segment_type_Offline             -3.5723      0.071    -50.107      0.000      -3.712      -3.433
market_segment_type_Online              -1.0307      0.048    -21.472      0.000      -1.125      -0.937
========================================================================================================

Observations

• Negative values of the coefficient shows that probability of cancelling the booking decreases with the increase of corresponding attribute value.

• Positive values of the coefficient show that that probability of cancelling the booking increases with the increase of corresponding attribute value.

• p-value of a variable indicates if the variable is significant or not. If we consider the significance level to be 0.05 (5%), then any variable with a p-value less than 0.05 would be considered significant.

Lets write a function that will build a loop and remove variables with p-value greater than 0.05 one by one then check the results again and repeat till we get no variables with p-values greater than 0.05.

# initial list of columns
cols = X_train4.columns.tolist()

# setting an initial max p-value
max_p_value = 1

while len(cols) > 0:
    # defining the train set
    x_train_aux = X_train4[cols]

    # fitting the model
    model = sm.OLS(y_train, x_train_aux).fit()

    # getting the p-values and the maximum p-value
    p_values = model.pvalues
    max_p_value = max(p_values)

    # name of the variable with maximum p-value
    feature_with_p_max = p_values.idxmax()

    if max_p_value > 0.05:
        cols.remove(feature_with_p_max)
    else:
        break

selected_features = cols
print(selected_features)

['no_of_children', 'no_of_week_nights', 'required_car_parking_space', 'lead_time', 'no_of_previous_bookings_not_canceled', 'no_of_special_requests', 'type_of_meal_plan_Meal Plan 2', 'type_of_meal_plan_Not Selected', 'room_type_reserved_Room_Type 2', 'room_type_reserved_Room_Type 4', 'room_type_reserved_Room_Type 5', 'room_type_reserved_Room_Type 6', 'room_type_reserved_Room_Type 7', 'market_segment_type_Complementary', 'market_segment_type_Corporate', 'market_segment_type_Offline', 'market_segment_type_Online']

X_train5 = X_train4[selected_features]
X_test5 = X_test4[selected_features]

X_train5 is the final Model with no variables with p-value greater than 0.05. Lets apply this model to statsmodels.

logit5 = sm.Logit(y_train, X_train5.astype(float))
lg5 = logit5.fit(
    disp=False
)  # setting disp=False will remove the information on number of iterations

print(lg5.summary())

                           Logit Regression Results                           
==============================================================================
Dep. Variable:         booking_status   No. Observations:                29733
Model:                          Logit   Df Residuals:                    29716
Method:                           MLE   Df Model:                           16
Date:                Fri, 22 Oct 2021   Pseudo R-squ.:                  0.3036
Time:                        20:28:44   Log-Likelihood:                -13251.
converged:                      False   LL-Null:                       -19028.
Covariance Type:            nonrobust   LLR p-value:                     0.000
========================================================================================================
                                           coef    std err          z      P>|z|      [0.025      0.975]
--------------------------------------------------------------------------------------------------------
no_of_children                           0.2457      0.047      5.178      0.000       0.153       0.339
no_of_week_nights                        0.0637      0.010      6.410      0.000       0.044       0.083
required_car_parking_space              -1.3611      0.113    -12.007      0.000      -1.583      -1.139
lead_time                                0.0161      0.000     64.748      0.000       0.016       0.017
no_of_previous_bookings_not_canceled    -1.3393      0.257     -5.214      0.000      -1.843      -0.836
no_of_special_requests                  -1.2421      0.023    -54.081      0.000      -1.287      -1.197
type_of_meal_plan_Meal Plan 2            0.3612      0.074      4.913      0.000       0.217       0.505
type_of_meal_plan_Not Selected           0.1597      0.041      3.913      0.000       0.080       0.240
room_type_reserved_Room_Type 2          -0.3779      0.123     -3.069      0.002      -0.619      -0.137
room_type_reserved_Room_Type 4           0.2896      0.039      7.434      0.000       0.213       0.366
room_type_reserved_Room_Type 5           0.5973      0.105      5.708      0.000       0.392       0.802
room_type_reserved_Room_Type 6           0.6117      0.111      5.528      0.000       0.395       0.829
room_type_reserved_Room_Type 7           0.7068      0.188      3.769      0.000       0.339       1.074
market_segment_type_Complementary      -28.7841   1.03e+05     -0.000      1.000   -2.02e+05    2.02e+05
market_segment_type_Corporate           -2.1452      0.110    -19.463      0.000      -2.361      -1.929
market_segment_type_Offline             -3.6447      0.066    -55.636      0.000      -3.773      -3.516
market_segment_type_Online              -1.1040      0.039    -28.019      0.000      -1.181      -1.027
========================================================================================================

Now no feature has p-value greater than 0.05, so we'll consider the features in X_train5 as the final ones and lg5 as final model.

Coefficient interpretations

• Coefficient of some variables positive an increase in these will lead to increase in chances cancellation of booking_status.

• Coefficient of some variables are negative increase in these will lead to decrease in chances cancellation of booking_status.

Converting coefficients to odds

• The coefficients of the logistic regression model are in terms of log(odd), to find the odds we have to take the exponential of the coefficients.
• Therefore, odds = exp(b)
• The percentage change in odds is given as odds = (exp(b) - 1) * 100

# converting coefficients to odds
odds = np.exp(lg5.params)

# finding the percentage change
perc_change_odds = (np.exp(lg5.params) - 1) * 100

# removing limit from number of columns to display
pd.set_option("display.max_columns", None)

# adding the odds to a dataframe
pd.DataFrame({"Odds": odds, "Change_odd%": perc_change_odds}, index=X_train5.columns).T

	no_of_children	no_of_week_nights	required_car_parking_space	lead_time	no_of_previous_bookings_not_canceled	no_of_special_requests	type_of_meal_plan_Meal Plan 2	type_of_meal_plan_Not Selected	room_type_reserved_Room_Type 2	room_type_reserved_Room_Type 4	room_type_reserved_Room_Type 5	room_type_reserved_Room_Type 6	room_type_reserved_Room_Type 7	market_segment_type_Complementary	market_segment_type_Corporate	market_segment_type_Offline	market_segment_type_Online
Odds	1.278538	1.065819	0.256368	1.016187	0.262038	0.288769	1.435031	1.173130	0.685332	1.335915	1.817174	1.843513	2.027485	3.156623e-13	0.117049	0.026130	0.331549
Change_odd%	27.853779	6.581891	-74.363197	1.618696	-73.796163	-71.123088	43.503120	17.313049	-31.466819	33.591539	81.717432	84.351304	102.748497	-1.000000e+02	-88.295098	-97.386957	-66.845080

Checking model performance on the training set

# creating confusion matrix
confusion_matrix_statsmodels(lg5, X_train5, y_train)

log_reg_model_train_perf = model_performance_classification_statsmodels(
    lg5, X_train5, y_train
)

print("Training performance:")
log_reg_model_train_perf

Training performance:

	Accuracy	Recall	Precision	F1
0	0.78169	0.58875	0.715718	0.646055

ROC-AUC

ROC-AUC on training set

logit_roc_auc_train = roc_auc_score(y_train, lg5.predict(X_train5))
fpr, tpr, thresholds = roc_curve(y_train, lg5.predict(X_train5))
plt.figure(figsize=(7, 5))
plt.plot(fpr, tpr, label="Logistic Regression (area = %0.2f)" % logit_roc_auc_train)
plt.plot([0, 1], [0, 1], "r--")
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel("False Positive Rate")
plt.ylabel("True Positive Rate")
plt.title("Receiver operating characteristic")
plt.legend(loc="lower right")
plt.show()

• Logistic Regression model is a low recall and ROC-AUC score.

Model Performance Improvement

Let's see if the recall score can be improved further, by changing the model threshold using AUC-ROC Curve.

Optimal threshold using AUC-ROC curve

# Optimal threshold as per AUC-ROC curve
# The optimal cut off would be where tpr is high and fpr is low
fpr, tpr, thresholds = roc_curve(y_train, lg5.predict(X_train5))

optimal_idx = np.argmax(tpr - fpr)
optimal_threshold_auc_roc = thresholds[optimal_idx]
print(optimal_threshold_auc_roc)

0.3345983648005056

Checking model performance on training set

# creating confusion matrix
confusion_matrix_statsmodels(
    lg5, X_train5, y_train, threshold=optimal_threshold_auc_roc
)

# checking model performance for this model
log_reg_model_train_perf_threshold_auc_roc = model_performance_classification_statsmodels(
    lg5, X_train5, y_train, threshold=optimal_threshold_auc_roc
)
print("Training performance:")
log_reg_model_train_perf_threshold_auc_roc

Training performance:

	Accuracy	Recall	Precision	F1
0	0.770255	0.781853	0.629209	0.697275

• Model performance has improved significantly.
• Model is giving a recall of 0.78 as compared to initial model which was giving a recall of 0.59.
• Precision has decreased from 0.71 to 0.62.

Let's use Precision-Recall curve and see if we can find a better threshold

y_scores = lg5.predict(X_train5)
prec, rec, tre = precision_recall_curve(y_train, y_scores,)


def plot_prec_recall_vs_tresh(precisions, recalls, thresholds):
    plt.plot(thresholds, precisions[:-1], "b--", label="precision")
    plt.plot(thresholds, recalls[:-1], "g--", label="recall")
    plt.xlabel("Threshold")
    plt.legend(loc="upper left")
    plt.ylim([0, 1])


plt.figure(figsize=(10, 7))
plot_prec_recall_vs_tresh(prec, rec, tre)
plt.show()

# setting the threshold
optimal_threshold_curve = 0.42

Checking model performance on training set

# creating confusion matrix
confusion_matrix_statsmodels(lg5, X_train5, y_train, threshold=optimal_threshold_curve)

log_reg_model_train_perf_threshold_curve = model_performance_classification_statsmodels(
    lg5, X_train5, y_train, threshold=optimal_threshold_curve
)
print("Training performance:")
log_reg_model_train_perf_threshold_curve

Training performance:

	Accuracy	Recall	Precision	F1
0	0.784785	0.685649	0.68071	0.683171

• Recall has reduced as compared to the initial model.
• Model is giving a better performance with 0.33 threshold found using AUC-ROC curve.

Model Performance Summary

# training performance comparison

models_train_comp_df = pd.concat(
    [
        log_reg_model_train_perf.T,
        log_reg_model_train_perf_threshold_auc_roc.T,
        log_reg_model_train_perf_threshold_curve.T,
    ],
    axis=1,
)
models_train_comp_df.columns = [
    "Logistic Regression sklearn",
    "Logistic Regression-0.33 Threshold",
    "Logistic Regression-0.42 Threshold",
]

print("Training performance comparison:")
models_train_comp_df

Training performance comparison:

	Logistic Regression sklearn	Logistic Regression-0.33 Threshold	Logistic Regression-0.42 Threshold
Accuracy	0.781690	0.770255	0.784785
Recall	0.588750	0.781853	0.685649
Precision	0.715718	0.629209	0.680710
F1	0.646055	0.697275	0.683171

Let's check the performance on the test set

Using model with default threshold

# creating confusion matrix
confusion_matrix_statsmodels(lg5, X_test5, y_test)

log_reg_model_test_perf = model_performance_classification_statsmodels(
    lg5, X_test5, y_test
)

print("Test performance:")
log_reg_model_test_perf

Test performance:

	Accuracy	Recall	Precision	F1
0	0.780681	0.592358	0.72536	0.652147

ROC curve on test set

logit_roc_auc_train = roc_auc_score(y_test, lg5.predict(X_test5))
fpr, tpr, thresholds = roc_curve(y_test, lg5.predict(X_test5))
plt.figure(figsize=(7, 5))
plt.plot(fpr, tpr, label="Logistic Regression (area = %0.2f)" % logit_roc_auc_train)
plt.plot([0, 1], [0, 1], "r--")
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel("False Positive Rate")
plt.ylabel("True Positive Rate")
plt.title("Receiver operating characteristic")
plt.legend(loc="lower right")
plt.show()

Using model with threshold=0.33

# creating confusion matrix
confusion_matrix_statsmodels(lg5, X_test5, y_test, threshold=optimal_threshold_auc_roc)

# checking model performance for this model
log_reg_model_test_perf_threshold_auc_roc = model_performance_classification_statsmodels(
    lg5, X_test5, y_test, threshold=optimal_threshold_auc_roc
)
print("Test performance:")
log_reg_model_test_perf_threshold_auc_roc

Test performance:

	Accuracy	Recall	Precision	F1
0	0.7664	0.781144	0.632321	0.698898

Using model with threshold = 0.42

# creating confusion matrix
confusion_matrix_statsmodels(lg5, X_test5, y_test, threshold=optimal_threshold_curve)

log_reg_model_test_perf_threshold_curve = model_performance_classification_statsmodels(
    lg5, X_test5, y_test, threshold=optimal_threshold_curve
)
print("Test performance:")
log_reg_model_test_perf_threshold_curve

Test performance:

	Accuracy	Recall	Precision	F1
0	0.780289	0.678725	0.685232	0.681963

Model performance summary

# training performance comparison

models_train_comp_df = pd.concat(
    [
        log_reg_model_train_perf.T,
        log_reg_model_train_perf_threshold_auc_roc.T,
        log_reg_model_train_perf_threshold_curve.T,
    ],
    axis=1,
)
models_train_comp_df.columns = [
    "Logistic Regression sklearn",
    "Logistic Regression-0.33 Threshold",
    "Logistic Regression-0.42 Threshold",
]

print("Training performance comparison:")
models_train_comp_df

Training performance comparison:

	Logistic Regression sklearn	Logistic Regression-0.33 Threshold	Logistic Regression-0.42 Threshold
Accuracy	0.781690	0.770255	0.784785
Recall	0.588750	0.781853	0.685649
Precision	0.715718	0.629209	0.680710
F1	0.646055	0.697275	0.683171

# testing performance comparison

models_test_comp_df = pd.concat(
    [
        log_reg_model_test_perf.T,
        log_reg_model_test_perf_threshold_auc_roc.T,
        log_reg_model_test_perf_threshold_curve.T,
    ],
    axis=1,
)
models_test_comp_df.columns = [
    "Logistic Regression sklearn",
    "Logistic Regression-0.33 Threshold",
    "Logistic Regression-0.42 Threshold",
]

print("Test set performance comparison:")
models_test_comp_df

Test set performance comparison:

	Logistic Regression sklearn	Logistic Regression-0.33 Threshold	Logistic Regression-0.42 Threshold
Accuracy	0.780681	0.766400	0.780289
Recall	0.592358	0.781144	0.678725
Precision	0.725360	0.632321	0.685232
F1	0.652147	0.698898	0.681963

Conclusion

• All the models are giving a generalized performance on training and test set.
• The highest recall is ~78% on the training set.
• Using the model with default threshold the model will give a low recall but good precision scores - This model will help the Hotel save resources on marketing.
• Using the model with 0.33 threshold the model will give a high recall but low precision scores - This model will help the Hotel identify customers will cancel the bookings effectively but the cost of resources will be high.
• Using the model with 0.42 threshold the model will give a balance recall and precision score - This model will help the Hotel to maintain a balance in identifying potential cancellations and the cost of resources.

Building a Decision Tree model

• We will build our model using the DecisionTreeClassifier function. Using default 'gini' criteria to split.
• class_weight is a hyperparameter for the decision tree classifier.

X = dummy_data.drop("booking_status", axis=1)  # Features
y = dummy_data["booking_status"]  # Labels (Target Variable)

# Splitting data into training and test set:
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1)
print(X_train.shape, X_test.shape)

(29733, 27) (12744, 27)

print("Number of rows in train data =", X_train.shape[0])
print("Number of rows in test data =", X_test.shape[0])

Number of rows in train data = 29733
Number of rows in test data = 12744

print("Percentage of classes in training set:")
print(y_train.value_counts(normalize=True))
print("Percentage of classes in test set:")
print(y_test.value_counts(normalize=True))

Percentage of classes in training set:
0    0.661588
1    0.338412
Name: booking_status, dtype: float64
Percentage of classes in test set:
0    0.652935
1    0.347065
Name: booking_status, dtype: float64

model = DecisionTreeClassifier(criterion="gini", random_state=1)
model.fit(X_train, y_train)

DecisionTreeClassifier(random_state=1)

Checking model performance on training set

decision_tree_perf_train = model_performance_classification_statsmodels(
    model, X_train, y_train
)
decision_tree_perf_train

	Accuracy	Recall	Precision	F1
0	0.9962	0.98877	1.0	0.994353

confusion_matrix_statsmodels(model, X_train, y_train)

Checking model performance on test set

decision_tree_perf_test = model_performance_classification_statsmodels(
    model, X_test, y_test
)
decision_tree_perf_test

	Accuracy	Recall	Precision	F1
0	0.793393	0.70269	0.702214	0.702452

confusion_matrix_statsmodels(model, X_test, y_test)

There is a huge disparity in performance of model on training set and test set, which suggests that the model is overfiiting.

Visualizing the Decision Tree

## creating a list of column names
feature_names = X_train.columns.to_list()

plt.figure(figsize=(20, 30))
out = tree.plot_tree(
    model,
    feature_names=feature_names,
    filled=True,
    fontsize=9,
    node_ids=False,
    class_names=None,
)
# below code will add arrows to the decision tree split if they are missing
for o in out:
    arrow = o.arrow_patch
    if arrow is not None:
        arrow.set_edgecolor("black")
        arrow.set_linewidth(1)
plt.show()

It is a very complex tree and difficult to interpret

# Text report showing the rules of a decision tree -

print(tree.export_text(model, feature_names=feature_names, show_weights=True))

|--- lead_time <= 150.50
|   |--- no_of_special_requests <= 0.50
|   |   |--- market_segment_type_Online <= 0.50
|   |   |   |--- lead_time <= 86.50
|   |   |   |   |--- avg_price_per_room <= 202.50
|   |   |   |   |   |--- no_of_weekend_nights <= 3.50
|   |   |   |   |   |   |--- repeated_guest <= 0.50
|   |   |   |   |   |   |   |--- market_segment_type_Offline <= 0.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 59.50
|   |   |   |   |   |   |   |   |   |--- weights: [102.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  59.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 63.20
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 9.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  9.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  63.20
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 43.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 27
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  43.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |--- market_segment_type_Offline >  0.50
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [738.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 9.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 41.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  41.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |--- arrival_month >  9.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 65.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  65.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |--- repeated_guest >  0.50
|   |   |   |   |   |   |   |--- lead_time <= 33.00
|   |   |   |   |   |   |   |   |--- weights: [354.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- lead_time >  33.00
|   |   |   |   |   |   |   |   |--- no_of_previous_bookings_not_canceled <= 10.50
|   |   |   |   |   |   |   |   |   |--- weights: [25.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_previous_bookings_not_canceled >  10.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 13.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_date >  13.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- no_of_weekend_nights >  3.50
|   |   |   |   |   |   |--- avg_price_per_room <= 92.72
|   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |--- avg_price_per_room >  92.72
|   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |--- avg_price_per_room >  202.50
|   |   |   |   |   |--- arrival_date <= 29.00
|   |   |   |   |   |   |--- weights: [0.00, 7.00] class: 1
|   |   |   |   |   |--- arrival_date >  29.00
|   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |--- lead_time >  86.50
|   |   |   |   |--- avg_price_per_room <= 93.33
|   |   |   |   |   |--- lead_time <= 133.50
|   |   |   |   |   |   |--- lead_time <= 132.50
|   |   |   |   |   |   |   |--- no_of_week_nights <= 6.50
|   |   |   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 58.75
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 124.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [28.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  124.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  58.75
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 61.12
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  61.12
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |   |   |--- weights: [27.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_week_nights >  6.50
|   |   |   |   |   |   |   |   |--- lead_time <= 110.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- lead_time >  110.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |--- lead_time >  132.50
|   |   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |--- lead_time >  133.50
|   |   |   |   |   |   |--- arrival_date <= 6.50
|   |   |   |   |   |   |   |--- arrival_month <= 7.50
|   |   |   |   |   |   |   |   |--- lead_time <= 136.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- lead_time >  136.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |--- arrival_month >  7.50
|   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |--- arrival_date >  6.50
|   |   |   |   |   |   |   |--- arrival_date <= 11.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 73.62
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  73.62
|   |   |   |   |   |   |   |   |   |--- weights: [12.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_date >  11.50
|   |   |   |   |   |   |   |   |--- weights: [61.00, 0.00] class: 0
|   |   |   |   |--- avg_price_per_room >  93.33
|   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |--- arrival_date <= 7.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_date >  7.50
|   |   |   |   |   |   |   |   |--- lead_time <= 101.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |--- lead_time >  101.00
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |--- avg_price_per_room <= 116.50
|   |   |   |   |   |   |   |   |--- lead_time <= 106.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 99.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  99.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- lead_time >  106.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 18.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 11.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  11.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- arrival_date >  18.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 23.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  23.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |--- avg_price_per_room >  116.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |--- avg_price_per_room <= 180.00
|   |   |   |   |   |   |   |--- avg_price_per_room <= 95.02
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 30.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |   |   |   |--- arrival_date >  30.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- avg_price_per_room >  95.02
|   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |--- arrival_year <= 2018.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 105.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  105.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- arrival_year >  2018.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 95.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  95.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 6.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 144.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  144.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- arrival_month >  6.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 115.55
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  115.55
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |--- avg_price_per_room >  180.00
|   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |--- market_segment_type_Online >  0.50
|   |   |   |--- lead_time <= 9.50
|   |   |   |   |--- avg_price_per_room <= 200.38
|   |   |   |   |   |--- lead_time <= 2.50
|   |   |   |   |   |   |--- no_of_week_nights <= 8.50
|   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |--- weights: [102.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |--- arrival_month <= 8.50
|   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 77.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  77.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 78.77
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [60.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  78.77
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 18
|   |   |   |   |   |   |   |   |--- arrival_month >  8.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 22.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 9.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  9.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [72.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_date >  22.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |--- no_of_week_nights >  8.50
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- lead_time >  2.50
|   |   |   |   |   |   |--- avg_price_per_room <= 99.42
|   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |--- weights: [75.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 83.01
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 5.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [47.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  5.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  83.01
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 21.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  21.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [17.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 77.70
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  77.70
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |--- avg_price_per_room >  99.42
|   |   |   |   |   |   |   |--- arrival_month <= 9.50
|   |   |   |   |   |   |   |   |--- arrival_year <= 2018.50
|   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |--- arrival_year >  2018.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 22.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  22.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_month >  9.50
|   |   |   |   |   |   |   |   |--- lead_time <= 7.50
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [40.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 11.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  11.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [8.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- lead_time >  7.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |--- avg_price_per_room >  200.38
|   |   |   |   |   |--- arrival_month <= 2.00
|   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- arrival_month >  2.00
|   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |--- weights: [0.00, 59.00] class: 1
|   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |--- lead_time >  9.50
|   |   |   |   |--- avg_price_per_room <= 200.35
|   |   |   |   |   |--- required_car_parking_space <= 0.50
|   |   |   |   |   |   |--- avg_price_per_room <= 105.28
|   |   |   |   |   |   |   |--- lead_time <= 25.50
|   |   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |   |--- weights: [79.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 20
|   |   |   |   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [67.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- lead_time >  25.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 63.11
|   |   |   |   |   |   |   |   |   |--- lead_time <= 69.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 27.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  27.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- lead_time >  69.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  63.11
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 7.50
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 22
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 24
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  7.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 24.00] class: 1
|   |   |   |   |   |   |--- avg_price_per_room >  105.28
|   |   |   |   |   |   |   |--- arrival_year <= 2018.50
|   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 94.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 10.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  10.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- lead_time >  94.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 10.50
|   |   |   |   |   |   |   |   |   |   |--- room_type_reserved_Room_Type 5 <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 26
|   |   |   |   |   |   |   |   |   |   |--- room_type_reserved_Room_Type 5 >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- arrival_month >  10.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 35.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  35.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |--- arrival_year >  2018.50
|   |   |   |   |   |   |   |   |--- lead_time <= 24.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_children <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- no_of_children >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- lead_time >  24.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 134.08
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 18
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  134.08
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 18
|   |   |   |   |   |--- required_car_parking_space >  0.50
|   |   |   |   |   |   |--- no_of_week_nights <= 9.00
|   |   |   |   |   |   |   |--- weights: [93.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_week_nights >  9.00
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |--- avg_price_per_room >  200.35
|   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |--- weights: [0.00, 264.00] class: 1
|   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |--- no_of_special_requests >  0.50
|   |   |--- no_of_special_requests <= 1.50
|   |   |   |--- lead_time <= 7.50
|   |   |   |   |--- no_of_week_nights <= 9.50
|   |   |   |   |   |--- avg_price_per_room <= 140.62
|   |   |   |   |   |   |--- no_of_adults <= 0.50
|   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- no_of_adults >  0.50
|   |   |   |   |   |   |   |--- avg_price_per_room <= 74.75
|   |   |   |   |   |   |   |   |--- lead_time <= 6.50
|   |   |   |   |   |   |   |   |   |--- weights: [319.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- lead_time >  6.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 3.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 14.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  14.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- arrival_month >  3.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [14.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- avg_price_per_room >  74.75
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 3.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_children <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |   |--- no_of_children >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- lead_time >  2.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 13.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  13.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  3.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 1.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  1.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- avg_price_per_room >  140.62
|   |   |   |   |   |   |--- lead_time <= 4.50
|   |   |   |   |   |   |   |--- avg_price_per_room <= 241.00
|   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 142.10
|   |   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  142.10
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 179.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  179.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |--- avg_price_per_room >  241.00
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 243.40
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  243.40
|   |   |   |   |   |   |   |   |   |--- lead_time <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- lead_time >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [6.00, 0.00] class: 0
|   |   |   |   |   |   |--- lead_time >  4.50
|   |   |   |   |   |   |   |--- arrival_date <= 23.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 143.75
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 141.67
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  141.67
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  143.75
|   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 5.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [22.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  5.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |--- arrival_date >  23.50
|   |   |   |   |   |   |   |   |--- weights: [26.00, 0.00] class: 0
|   |   |   |   |--- no_of_week_nights >  9.50
|   |   |   |   |   |--- arrival_month <= 7.50
|   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |--- arrival_month >  7.50
|   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |--- lead_time >  7.50
|   |   |   |   |--- avg_price_per_room <= 121.78
|   |   |   |   |   |--- market_segment_type_Online <= 0.50
|   |   |   |   |   |   |--- lead_time <= 94.50
|   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [487.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected >  0.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 99.00
|   |   |   |   |   |   |   |   |   |--- weights: [20.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  99.00
|   |   |   |   |   |   |   |   |   |--- lead_time <= 63.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  63.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |--- lead_time >  94.50
|   |   |   |   |   |   |   |--- no_of_weekend_nights <= 3.00
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 86.60
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 25.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  25.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 147.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  147.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  86.60
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 89.55
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 5.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  5.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  89.55
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |--- no_of_weekend_nights >  3.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- market_segment_type_Online >  0.50
|   |   |   |   |   |   |--- lead_time <= 90.50
|   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |--- no_of_week_nights <= 9.50
|   |   |   |   |   |   |   |   |   |--- weights: [269.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_week_nights >  9.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 33.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  33.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 2.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 67.36
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  67.36
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 28
|   |   |   |   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [233.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  2.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 8.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 5.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  5.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  8.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 10.00] class: 1
|   |   |   |   |   |   |--- lead_time >  90.50
|   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |--- lead_time <= 96.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 11.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 8.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  8.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_date >  11.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 9.00] class: 1
|   |   |   |   |   |   |   |   |--- lead_time >  96.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 24.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 7.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  7.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- arrival_date >  24.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 51.23
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  51.23
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [10.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 72.08
|   |   |   |   |   |   |   |   |   |--- weights: [46.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  72.08
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 3.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 127.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  127.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |--- arrival_month >  3.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 9.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  9.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |--- avg_price_per_room >  121.78
|   |   |   |   |   |--- required_car_parking_space <= 0.50
|   |   |   |   |   |   |--- arrival_year <= 2018.50
|   |   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |   |--- arrival_month <= 8.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 7.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 19.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 18
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  19.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  7.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_month >  8.50
|   |   |   |   |   |   |   |   |   |--- room_type_reserved_Room_Type 4 <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_children <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- no_of_children >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |--- room_type_reserved_Room_Type 4 >  0.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 162.70
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 21
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  162.70
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |   |--- lead_time <= 102.50
|   |   |   |   |   |   |   |   |   |--- weights: [68.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- lead_time >  102.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |--- arrival_year >  2018.50
|   |   |   |   |   |   |   |--- arrival_month <= 7.50
|   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected <= 0.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 177.75
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 117.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 22
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  117.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  177.75
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 93.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  93.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected >  0.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 66.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 30.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 20
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  30.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  66.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 134.17
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  134.17
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_month >  7.50
|   |   |   |   |   |   |   |   |--- lead_time <= 134.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 14.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 8.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  8.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- lead_time >  14.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 33.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  33.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 17
|   |   |   |   |   |   |   |   |--- lead_time >  134.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 11.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 2.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  2.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [11.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_date >  11.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 139.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  139.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |--- required_car_parking_space >  0.50
|   |   |   |   |   |   |--- weights: [129.00, 0.00] class: 0
|   |   |--- no_of_special_requests >  1.50
|   |   |   |--- lead_time <= 90.50
|   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |--- weights: [2973.00, 0.00] class: 0
|   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |--- no_of_week_nights <= 9.50
|   |   |   |   |   |   |--- no_of_special_requests <= 2.50
|   |   |   |   |   |   |   |--- lead_time <= 8.50
|   |   |   |   |   |   |   |   |--- no_of_week_nights <= 4.50
|   |   |   |   |   |   |   |   |   |--- weights: [51.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_week_nights >  4.50
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 6.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  6.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [18.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- lead_time >  8.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 30.50
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [30.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 91.10
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  91.10
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 16
|   |   |   |   |   |   |   |   |--- arrival_date >  30.50
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 77.09
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  77.09
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_special_requests >  2.50
|   |   |   |   |   |   |   |--- weights: [97.00, 0.00] class: 0
|   |   |   |   |   |--- no_of_week_nights >  9.50
|   |   |   |   |   |   |--- no_of_special_requests <= 2.50
|   |   |   |   |   |   |   |--- no_of_previous_bookings_not_canceled <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 7.00] class: 1
|   |   |   |   |   |   |   |--- no_of_previous_bookings_not_canceled >  0.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_special_requests >  2.50
|   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |--- lead_time >  90.50
|   |   |   |   |--- avg_price_per_room <= 200.70
|   |   |   |   |   |--- no_of_special_requests <= 2.50
|   |   |   |   |   |   |--- arrival_year <= 2018.50
|   |   |   |   |   |   |   |--- arrival_month <= 8.50
|   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 14.00
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 7.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  7.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- arrival_date >  14.00
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 6.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 2.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  2.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  6.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 25.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  25.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_month >  8.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 8.50
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 118.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  118.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_date >  8.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 11.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 104.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  104.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- arrival_date >  11.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 144.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  144.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |--- arrival_year >  2018.50
|   |   |   |   |   |   |   |--- avg_price_per_room <= 95.25
|   |   |   |   |   |   |   |   |--- no_of_week_nights <= 4.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 74.82
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 99.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  99.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  74.82
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |--- no_of_week_nights >  4.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |   |--- avg_price_per_room >  95.25
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 172.12
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 5.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  5.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  172.12
|   |   |   |   |   |   |   |   |   |--- lead_time <= 145.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 141.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  141.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  145.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |--- no_of_special_requests >  2.50
|   |   |   |   |   |   |--- weights: [167.00, 0.00] class: 0
|   |   |   |   |--- avg_price_per_room >  200.70
|   |   |   |   |   |--- no_of_special_requests <= 2.50
|   |   |   |   |   |   |--- weights: [0.00, 23.00] class: 1
|   |   |   |   |   |--- no_of_special_requests >  2.50
|   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|--- lead_time >  150.50
|   |--- avg_price_per_room <= 100.04
|   |   |--- market_segment_type_Online <= 0.50
|   |   |   |--- lead_time <= 348.50
|   |   |   |   |--- avg_price_per_room <= 89.76
|   |   |   |   |   |--- no_of_special_requests <= 0.50
|   |   |   |   |   |   |--- lead_time <= 258.00
|   |   |   |   |   |   |   |--- no_of_previous_bookings_not_canceled <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 3.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 25.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 159.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  159.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |--- arrival_date >  25.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  3.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- no_of_previous_bookings_not_canceled >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- lead_time >  258.00
|   |   |   |   |   |   |   |--- avg_price_per_room <= 88.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 20.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 6.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_date >  6.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_year <= 2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- arrival_year >  2017.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |--- arrival_date >  20.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 84.38
|   |   |   |   |   |   |   |   |   |   |--- weights: [14.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  84.38
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- avg_price_per_room >  88.50
|   |   |   |   |   |   |   |   |--- lead_time <= 328.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 10.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |--- arrival_date >  10.00
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 309.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  309.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- lead_time >  328.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- no_of_special_requests >  0.50
|   |   |   |   |   |   |--- lead_time <= 151.50
|   |   |   |   |   |   |   |--- arrival_month <= 7.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |--- arrival_month >  7.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- lead_time >  151.50
|   |   |   |   |   |   |   |--- arrival_date <= 30.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 3.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 4.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [9.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  4.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_date >  3.50
|   |   |   |   |   |   |   |   |   |--- weights: [97.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_date >  30.50
|   |   |   |   |   |   |   |   |--- lead_time <= 168.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- lead_time >  168.00
|   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |--- avg_price_per_room >  89.76
|   |   |   |   |   |--- no_of_special_requests <= 0.50
|   |   |   |   |   |   |--- arrival_month <= 6.50
|   |   |   |   |   |   |   |--- arrival_date <= 2.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_date >  2.50
|   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 7.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  7.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.00] class: 1
|   |   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 243.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- lead_time >  243.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |--- arrival_month >  6.50
|   |   |   |   |   |   |   |--- lead_time <= 233.50
|   |   |   |   |   |   |   |   |--- lead_time <= 170.50
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 3.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  3.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 6.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  6.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- lead_time >  170.50
|   |   |   |   |   |   |   |   |   |--- weights: [12.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- lead_time >  233.50
|   |   |   |   |   |   |   |   |--- no_of_adults <= 1.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 98.00
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 310.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  310.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  98.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_adults >  1.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 95.47
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  95.47
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 7.00] class: 1
|   |   |   |   |   |--- no_of_special_requests >  0.50
|   |   |   |   |   |   |--- avg_price_per_room <= 90.25
|   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_special_requests <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_special_requests >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |--- avg_price_per_room >  90.25
|   |   |   |   |   |   |   |--- arrival_date <= 30.00
|   |   |   |   |   |   |   |   |--- arrival_date <= 10.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 8.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [9.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_date >  8.50
|   |   |   |   |   |   |   |   |   |   |--- lead_time <= 189.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- lead_time >  189.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_date >  10.50
|   |   |   |   |   |   |   |   |   |--- weights: [21.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- arrival_date >  30.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |--- lead_time >  348.50
|   |   |   |   |--- no_of_special_requests <= 0.50
|   |   |   |   |   |--- arrival_month <= 9.50
|   |   |   |   |   |   |--- avg_price_per_room <= 91.42
|   |   |   |   |   |   |   |--- weights: [0.00, 34.00] class: 1
|   |   |   |   |   |   |--- avg_price_per_room >  91.42
|   |   |   |   |   |   |   |--- lead_time <= 414.00
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- lead_time >  414.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- arrival_month >  9.50
|   |   |   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |   |   |--- arrival_date <= 13.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 93.33
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  93.33
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |--- arrival_date >  13.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |--- no_of_special_requests >  0.50
|   |   |   |   |   |--- avg_price_per_room <= 73.44
|   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |--- avg_price_per_room >  73.44
|   |   |   |   |   |   |--- arrival_date <= 9.00
|   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |--- arrival_date >  9.00
|   |   |   |   |   |   |   |--- avg_price_per_room <= 92.75
|   |   |   |   |   |   |   |   |--- no_of_special_requests <= 2.00
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 1.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  1.00
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 84.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  84.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_special_requests >  2.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- avg_price_per_room >  92.75
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |--- market_segment_type_Online >  0.50
|   |   |   |--- no_of_special_requests <= 0.50
|   |   |   |   |--- avg_price_per_room <= 2.50
|   |   |   |   |   |--- lead_time <= 285.50
|   |   |   |   |   |   |--- arrival_date <= 8.50
|   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- type_of_meal_plan_Meal Plan 2 >  0.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 1.00] class: 0
|   |   |   |   |   |   |--- arrival_date >  8.50
|   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |--- lead_time >  285.50
|   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |--- avg_price_per_room >  2.50
|   |   |   |   |   |--- no_of_adults <= 2.50
|   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |--- weights: [0.00, 595.00] class: 1
|   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 72.27
|   |   |   |   |   |   |   |   |   |--- lead_time <= 277.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- lead_time >  277.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  72.27
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |--- lead_time <= 223.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 212.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |   |   |   |--- lead_time >  212.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_children <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_children >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- lead_time >  223.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 38.00] class: 1
|   |   |   |   |   |--- no_of_adults >  2.50
|   |   |   |   |   |   |--- avg_price_per_room <= 88.41
|   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |--- avg_price_per_room >  88.41
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |--- no_of_special_requests >  0.50
|   |   |   |   |--- no_of_weekend_nights <= 0.50
|   |   |   |   |   |--- lead_time <= 180.50
|   |   |   |   |   |   |--- arrival_month <= 2.50
|   |   |   |   |   |   |   |--- avg_price_per_room <= 88.55
|   |   |   |   |   |   |   |   |--- lead_time <= 172.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |--- lead_time >  172.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- avg_price_per_room >  88.55
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |--- arrival_month >  2.50
|   |   |   |   |   |   |   |--- no_of_special_requests <= 1.50
|   |   |   |   |   |   |   |   |--- arrival_month <= 10.50
|   |   |   |   |   |   |   |   |   |--- arrival_date <= 8.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 6.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  6.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- arrival_date >  8.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 30.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  30.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_month >  10.50
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |--- no_of_special_requests >  1.50
|   |   |   |   |   |   |   |   |--- weights: [23.00, 0.00] class: 0
|   |   |   |   |   |--- lead_time >  180.50
|   |   |   |   |   |   |--- no_of_special_requests <= 2.50
|   |   |   |   |   |   |   |--- arrival_month <= 11.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 36.16
|   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  36.16
|   |   |   |   |   |   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 182.00] class: 1
|   |   |   |   |   |   |   |--- arrival_month >  11.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 6.50
|   |   |   |   |   |   |   |   |   |--- room_type_reserved_Room_Type 4 <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |--- room_type_reserved_Room_Type 4 >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- arrival_date >  6.50
|   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_children <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_children >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 4.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  4.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_special_requests >  2.50
|   |   |   |   |   |   |   |--- weights: [18.00, 0.00] class: 0
|   |   |   |   |--- no_of_weekend_nights >  0.50
|   |   |   |   |   |--- no_of_special_requests <= 2.50
|   |   |   |   |   |   |--- arrival_year <= 2018.50
|   |   |   |   |   |   |   |--- arrival_month <= 10.50
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 93.01
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 81.12
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  81.12
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |--- no_of_weekend_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 6.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  6.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  93.01
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 95.97
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  95.97
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |   |--- type_of_meal_plan_Not Selected >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |--- arrival_month >  10.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 29.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 306.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights <= 4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- no_of_week_nights >  4.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- lead_time >  306.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |--- arrival_date >  29.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |--- arrival_year >  2018.50
|   |   |   |   |   |   |   |--- avg_price_per_room <= 76.79
|   |   |   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |   |   |--- weights: [18.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |   |   |--- lead_time <= 199.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults <= 1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults >  1.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |--- lead_time >  199.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- avg_price_per_room >  76.79
|   |   |   |   |   |   |   |   |--- avg_price_per_room <= 78.06
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 7.00] class: 1
|   |   |   |   |   |   |   |   |--- avg_price_per_room >  78.06
|   |   |   |   |   |   |   |   |   |--- lead_time <= 330.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_date <= 8.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- arrival_date >  8.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |--- lead_time >  330.50
|   |   |   |   |   |   |   |   |   |   |--- arrival_month <= 6.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- arrival_month >  6.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |--- no_of_special_requests >  2.50
|   |   |   |   |   |   |--- weights: [42.00, 0.00] class: 0
|   |--- avg_price_per_room >  100.04
|   |   |--- no_of_special_requests <= 2.50
|   |   |   |--- arrival_month <= 11.50
|   |   |   |   |--- arrival_month <= 1.50
|   |   |   |   |   |--- no_of_week_nights <= 2.50
|   |   |   |   |   |   |--- lead_time <= 213.00
|   |   |   |   |   |   |   |--- lead_time <= 204.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- lead_time >  204.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- lead_time >  213.00
|   |   |   |   |   |   |   |--- weights: [6.00, 0.00] class: 0
|   |   |   |   |   |--- no_of_week_nights >  2.50
|   |   |   |   |   |   |--- no_of_special_requests <= 0.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_special_requests >  0.50
|   |   |   |   |   |   |   |--- arrival_date <= 15.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 8.00] class: 1
|   |   |   |   |   |   |   |--- arrival_date >  15.00
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |--- arrival_month >  1.50
|   |   |   |   |   |--- weights: [0.00, 2630.00] class: 1
|   |   |   |--- arrival_month >  11.50
|   |   |   |   |--- no_of_special_requests <= 0.50
|   |   |   |   |   |--- weights: [38.00, 0.00] class: 0
|   |   |   |   |--- no_of_special_requests >  0.50
|   |   |   |   |   |--- arrival_date <= 24.00
|   |   |   |   |   |   |--- avg_price_per_room <= 156.82
|   |   |   |   |   |   |   |--- weights: [6.00, 0.00] class: 0
|   |   |   |   |   |   |--- avg_price_per_room >  156.82
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- arrival_date >  24.00
|   |   |   |   |   |   |--- no_of_week_nights <= 3.50
|   |   |   |   |   |   |   |--- no_of_children <= 0.50
|   |   |   |   |   |   |   |   |--- arrival_date <= 29.50
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room <= 119.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults <= 2.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_adults >  2.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- avg_price_per_room >  119.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- arrival_date >  29.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |--- no_of_children >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |--- no_of_week_nights >  3.50
|   |   |   |   |   |   |   |--- weights: [0.00, 9.00] class: 1
|   |   |--- no_of_special_requests >  2.50
|   |   |   |--- weights: [149.00, 0.00] class: 0

# importance of features in the tree building ( The importance of a feature is computed as the
# (normalized) total reduction of the criterion brought by that feature. It is also known as the Gini importance )

print(
    pd.DataFrame(
        model.feature_importances_, columns=["Imp"], index=X_train.columns
    ).sort_values(by="Imp", ascending=False)
)

                                           Imp
lead_time                             0.347163
avg_price_per_room                    0.155664
no_of_special_requests                0.100093
arrival_date                          0.095252
market_segment_type_Online            0.079360
arrival_month                         0.064333
no_of_week_nights                     0.054437
no_of_weekend_nights                  0.028943
arrival_year                          0.018988
no_of_adults                          0.015609
room_type_reserved_Room_Type 4        0.009020
type_of_meal_plan_Not Selected        0.008478
required_car_parking_space            0.006936
no_of_children                        0.006726
type_of_meal_plan_Meal Plan 2         0.003169
room_type_reserved_Room_Type 2        0.001930
room_type_reserved_Room_Type 5        0.001459
room_type_reserved_Room_Type 6        0.000812
market_segment_type_Offline           0.000721
market_segment_type_Corporate         0.000372
repeated_guest                        0.000282
no_of_previous_bookings_not_canceled  0.000142
no_of_previous_cancellations          0.000111
type_of_meal_plan_Meal Plan 3         0.000000
room_type_reserved_Room_Type 3        0.000000
room_type_reserved_Room_Type 7        0.000000
market_segment_type_Complementary     0.000000

importances = model.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(12, 12))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="violet", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

• According to the decision tree model, lead time the most important variable for predicting the booking_status
• The tree above is very complex, such a tree often overfits.

Reducing over fitting

Using GridSearch for Hyperparameter tuning of our tree model

• Hyperparameter tuning is also tricky in the sense that there is no direct way to calculate how a change in the hyperparameter value will reduce the loss of your model, so we usually resort to experimentation. i.e we'll use Grid search
• Grid search is a tuning technique that attempts to compute the optimum values of hyperparameters.
• It is an exhaustive search that is performed on a the specific parameter values of a model.
• The parameters of the estimator/model used to apply these methods are optimized by cross-validated grid-search over a parameter grid.

# Choose the type of classifier.
estimator = DecisionTreeClassifier(random_state=1, class_weight={0: 0.15, 1: 0.85})

# Grid of parameters to choose from
parameters = {
    "max_depth": [5, 10, 15, None],
    "criterion": ["entropy", "gini"],
    "splitter": ["best", "random"],
    "min_impurity_decrease": [0.00001, 0.0001, 0.01],
}

# Type of scoring used to compare parameter combinations
scorer = make_scorer(recall_score)

# Run the grid search
grid_obj = GridSearchCV(estimator, parameters, scoring=scorer, cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
estimator = grid_obj.best_estimator_

# Fit the best algorithm to the data.
estimator.fit(X_train, y_train)

DecisionTreeClassifier(class_weight={0: 0.15, 1: 0.85}, criterion='entropy',
                       max_depth=5, min_impurity_decrease=1e-05, random_state=1,
                       splitter='random')

Checking performance on training set

decision_tree_tune_perf_train = model_performance_classification_statsmodels(
    estimator, X_train, y_train
)
decision_tree_tune_perf_train

	Accuracy	Recall	Precision	F1
0	0.571621	0.966706	0.439559	0.60433

confusion_matrix_statsmodels(estimator, X_train, y_train)

• The Recall has improved on the training set as compared to the initial model.

Checking model performance on test set

decision_tree_tune_perf_test = model_performance_classification_statsmodels(
    estimator, X_test, y_test
)

decision_tree_tune_perf_test

	Accuracy	Recall	Precision	F1
0	0.574231	0.967217	0.447536	0.61193

confusion_matrix_statsmodels(estimator, X_test, y_test)

• The model is giving a generalized result now.

Visualizing the Decision Tree

plt.figure(figsize=(15, 12))

tree.plot_tree(
    estimator,
    feature_names=feature_names,
    filled=True,
    fontsize=9,
    node_ids=True,
    class_names=True,
)
plt.show()

# Text report showing the rules of a decision tree -

print(tree.export_text(estimator, feature_names=feature_names, show_weights=True))

|--- lead_time <= 40.78
|   |--- market_segment_type_Online <= 0.01
|   |   |--- repeated_guest <= 0.07
|   |   |   |--- market_segment_type_Offline <= 0.38
|   |   |   |   |--- market_segment_type_Complementary <= 0.60
|   |   |   |   |   |--- weights: [95.10, 88.40] class: 0
|   |   |   |   |--- market_segment_type_Complementary >  0.60
|   |   |   |   |   |--- weights: [26.25, 0.00] class: 0
|   |   |   |--- market_segment_type_Offline >  0.38
|   |   |   |   |--- arrival_month <= 10.06
|   |   |   |   |   |--- weights: [178.80, 41.65] class: 0
|   |   |   |   |--- arrival_month >  10.06
|   |   |   |   |   |--- weights: [34.95, 0.85] class: 0
|   |   |--- repeated_guest >  0.07
|   |   |   |--- lead_time <= 31.22
|   |   |   |   |--- weights: [116.10, 0.00] class: 0
|   |   |   |--- lead_time >  31.22
|   |   |   |   |--- no_of_previous_cancellations <= 2.97
|   |   |   |   |   |--- weights: [2.10, 0.00] class: 0
|   |   |   |   |--- no_of_previous_cancellations >  2.97
|   |   |   |   |   |--- weights: [0.00, 0.85] class: 1
|   |--- market_segment_type_Online >  0.01
|   |   |--- no_of_special_requests <= 1.84
|   |   |   |--- no_of_special_requests <= 0.59
|   |   |   |   |--- lead_time <= 23.15
|   |   |   |   |   |--- weights: [265.65, 650.25] class: 1
|   |   |   |   |--- lead_time >  23.15
|   |   |   |   |   |--- weights: [58.35, 522.75] class: 1
|   |   |   |--- no_of_special_requests >  0.59
|   |   |   |   |--- lead_time <= 4.85
|   |   |   |   |   |--- weights: [130.95, 29.75] class: 0
|   |   |   |   |--- lead_time >  4.85
|   |   |   |   |   |--- weights: [383.70, 572.05] class: 1
|   |   |--- no_of_special_requests >  1.84
|   |   |   |--- no_of_week_nights <= 2.26
|   |   |   |   |--- weights: [199.05, 0.00] class: 0
|   |   |   |--- no_of_week_nights >  2.26
|   |   |   |   |--- no_of_special_requests <= 2.37
|   |   |   |   |   |--- weights: [78.60, 22.95] class: 0
|   |   |   |   |--- no_of_special_requests >  2.37
|   |   |   |   |   |--- weights: [16.50, 0.00] class: 0
|--- lead_time >  40.78
|   |--- market_segment_type_Online <= 0.51
|   |   |--- lead_time <= 184.18
|   |   |   |--- lead_time <= 101.75
|   |   |   |   |--- lead_time <= 85.49
|   |   |   |   |   |--- weights: [145.20, 61.20] class: 0
|   |   |   |   |--- lead_time >  85.49
|   |   |   |   |   |--- weights: [39.60, 39.10] class: 0
|   |   |   |--- lead_time >  101.75
|   |   |   |   |--- avg_price_per_room <= 195.32
|   |   |   |   |   |--- weights: [109.95, 175.95] class: 1
|   |   |   |   |--- avg_price_per_room >  195.32
|   |   |   |   |   |--- weights: [0.00, 8.50] class: 1
|   |   |--- lead_time >  184.18
|   |   |   |--- avg_price_per_room <= 109.83
|   |   |   |   |--- avg_price_per_room <= 84.90
|   |   |   |   |   |--- weights: [23.85, 34.00] class: 1
|   |   |   |   |--- avg_price_per_room >  84.90
|   |   |   |   |   |--- weights: [12.60, 90.95] class: 1
|   |   |   |--- avg_price_per_room >  109.83
|   |   |   |   |--- no_of_special_requests <= 2.58
|   |   |   |   |   |--- weights: [1.20, 58.65] class: 1
|   |   |   |   |--- no_of_special_requests >  2.58
|   |   |   |   |   |--- weights: [0.30, 0.00] class: 0
|   |--- market_segment_type_Online >  0.51
|   |   |--- lead_time <= 254.70
|   |   |   |--- lead_time <= 101.53
|   |   |   |   |--- required_car_parking_space <= 0.05
|   |   |   |   |   |--- weights: [616.05, 1995.80] class: 1
|   |   |   |   |--- required_car_parking_space >  0.05
|   |   |   |   |   |--- weights: [25.80, 0.85] class: 0
|   |   |   |--- lead_time >  101.53
|   |   |   |   |--- no_of_special_requests <= 0.05
|   |   |   |   |   |--- weights: [84.75, 1865.75] class: 1
|   |   |   |   |--- no_of_special_requests >  0.05
|   |   |   |   |   |--- weights: [282.00, 1719.55] class: 1
|   |   |--- lead_time >  254.70
|   |   |   |--- no_of_special_requests <= 0.71
|   |   |   |   |--- no_of_adults <= 2.70
|   |   |   |   |   |--- weights: [1.80, 230.35] class: 1
|   |   |   |   |--- no_of_adults >  2.70
|   |   |   |   |   |--- weights: [0.90, 18.70] class: 1
|   |   |   |--- no_of_special_requests >  0.71
|   |   |   |   |--- no_of_special_requests <= 3.11
|   |   |   |   |   |--- weights: [19.50, 323.85] class: 1
|   |   |   |   |--- no_of_special_requests >  3.11
|   |   |   |   |   |--- weights: [1.05, 0.00] class: 0

• The Tree is now less complex and easy to interpret.
• The Tree depth ahs reduced to 6.

# importance of features in the tree building ( The importance of a feature is computed as the
# (normalized) total reduction of the 'criterion' brought by that feature. It is also known as the Gini importance )

print(
    pd.DataFrame(
        estimator.feature_importances_, columns=["Imp"], index=X_train.columns
    ).sort_values(by="Imp", ascending=False)
)

# Here we will see that importance of features has increased

                                           Imp
lead_time                             0.489695
no_of_special_requests                0.236461
market_segment_type_Online            0.183413
required_car_parking_space            0.020084
repeated_guest                        0.018261
no_of_week_nights                     0.014591
avg_price_per_room                    0.012618
market_segment_type_Offline           0.011329
market_segment_type_Complementary     0.009471
arrival_month                         0.002564
no_of_previous_cancellations          0.001067
no_of_adults                          0.000445
no_of_previous_bookings_not_canceled  0.000000
no_of_children                        0.000000
type_of_meal_plan_Meal Plan 2         0.000000
arrival_date                          0.000000
type_of_meal_plan_Not Selected        0.000000
room_type_reserved_Room_Type 2        0.000000
room_type_reserved_Room_Type 3        0.000000
room_type_reserved_Room_Type 4        0.000000
room_type_reserved_Room_Type 5        0.000000
room_type_reserved_Room_Type 6        0.000000
room_type_reserved_Room_Type 7        0.000000
arrival_year                          0.000000
market_segment_type_Corporate         0.000000
no_of_weekend_nights                  0.000000
type_of_meal_plan_Meal Plan 3         0.000000

importances = estimator.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(12, 12))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="violet", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

• Lead_time is still the most important feature for predicting the booking_status

Cost Complexity Pruning

The DecisionTreeClassifier provides parameters such as min_samples_leaf and max_depth to prevent a tree from overfiting. Cost complexity pruning provides another option to control the size of a tree. In DecisionTreeClassifier, this pruning technique is parameterized by the cost complexity parameter, ccp_alpha. Greater values of ccp_alpha increase the number of nodes pruned. Here we only show the effect of ccp_alpha on regularizing the trees and how to choose a ccp_alpha based on validation scores.

Total impurity of leaves vs effective alphas of pruned tree

Minimal cost complexity pruning recursively finds the node with the "weakest link". The weakest link is characterized by an effective alpha, where the nodes with the smallest effective alpha are pruned first. To get an idea of what values of ccp_alpha could be appropriate, scikit-learn provides DecisionTreeClassifier.cost_complexity_pruning_path that returns the effective alphas and the corresponding total leaf impurities at each step of the pruning process. As alpha increases, more of the tree is pruned, which increases the total impurity of its leaves.

clf = DecisionTreeClassifier(random_state=1)
path = clf.cost_complexity_pruning_path(X_train, y_train)
ccp_alphas, impurities = path.ccp_alphas, path.impurities

pd.DataFrame(path)

	ccp_alphas	impurities
0	0.000000	0.003800
1	0.000000	0.003800
2	0.000000	0.003800
3	0.000000	0.003800
4	0.000000	0.003800
...	...	...
2005	0.009093	0.299653
2006	0.012134	0.311787
2007	0.012991	0.324779
2008	0.024660	0.374098
2009	0.073681	0.447779

2010 rows × 2 columns

fig, ax = plt.subplots(figsize=(15, 5))
ax.plot(ccp_alphas[:-1], impurities[:-1], marker="o", drawstyle="steps-post")
ax.set_xlabel("effective alpha")
ax.set_ylabel("total impurity of leaves")
ax.set_title("Total Impurity vs effective alpha for training set")
plt.show()

Next, we train a decision tree using the effective alphas. The last value in ccp_alphas is the alpha value that prunes the whole tree, leaving the tree, clfs[-1], with one node.

clfs = []
for ccp_alpha in ccp_alphas:
    clf = DecisionTreeClassifier(random_state=1, ccp_alpha=ccp_alpha)
    clf.fit(X_train, y_train)
    clfs.append(clf)
print(
    "Number of nodes in the last tree is: {} with ccp_alpha: {}".format(
        clfs[-1].tree_.node_count, ccp_alphas[-1]
    )
)

Number of nodes in the last tree is: 1 with ccp_alpha: 0.07368068552560186

For the remainder, we remove the last element in clfs and ccp_alphas, because it is the trivial tree with only one node. Here we show that the number of nodes and tree depth decreases as alpha increases.

clfs = clfs[:-1]
ccp_alphas = ccp_alphas[:-1]

node_counts = [clf.tree_.node_count for clf in clfs]
depth = [clf.tree_.max_depth for clf in clfs]
fig, ax = plt.subplots(2, 1, figsize=(10, 7))
ax[0].plot(ccp_alphas, node_counts, marker="o", drawstyle="steps-post")
ax[0].set_xlabel("alpha")
ax[0].set_ylabel("number of nodes")
ax[0].set_title("Number of nodes vs alpha")
ax[1].plot(ccp_alphas, depth, marker="o", drawstyle="steps-post")
ax[1].set_xlabel("alpha")
ax[1].set_ylabel("depth of tree")
ax[1].set_title("Depth vs alpha")
fig.tight_layout()

Recall vs alpha for training and testing sets

recall_train = []
for clf in clfs:
    pred_train = clf.predict(X_train)
    values_train = recall_score(y_train, pred_train)
    recall_train.append(values_train)

recall_test = []
for clf in clfs:
    pred_test = clf.predict(X_test)
    values_test = recall_score(y_test, pred_test)
    recall_test.append(values_test)

fig, ax = plt.subplots(figsize=(15, 5))
ax.set_xlabel("alpha")
ax.set_ylabel("Recall")
ax.set_title("Recall vs alpha for training and testing sets")
ax.plot(ccp_alphas, recall_train, marker="o", label="train", drawstyle="steps-post")
ax.plot(ccp_alphas, recall_test, marker="o", label="test", drawstyle="steps-post")
ax.legend()
plt.show()

# creating the model where we get highest train and test recall
index_best_model = np.argmax(recall_test)
best_model = clfs[index_best_model]
print(best_model)

DecisionTreeClassifier(ccp_alpha=0.01299122169370226, random_state=1)

Checking model performance on training set

decision_tree_postpruned_perf_train = model_performance_classification_statsmodels(
    best_model, X_train, y_train
)
decision_tree_postpruned_perf_train

	Accuracy	Recall	Precision	F1
0	0.756903	0.762175	0.613324	0.679695

confusion_matrix_statsmodels(best_model, X_train, y_train)

Checking model performance on test set

decision_tree_postpruned_perf_test = model_performance_classification_statsmodels(
    best_model, X_test, y_test
)
decision_tree_postpruned_perf_test

	Accuracy	Recall	Precision	F1
0	0.755257	0.7669	0.618978	0.685045

confusion_matrix_statsmodels(best_model, X_train, y_train)

With post-pruning we are getting good and generalized model performance on both training and test set.

Visualizing the Decision Tree

plt.figure(figsize=(10, 10))

out = tree.plot_tree(
    best_model,
    feature_names=feature_names,
    filled=True,
    fontsize=9,
    node_ids=True,
    class_names=True,
)
for o in out:
    arrow = o.arrow_patch
    if arrow is not None:
        arrow.set_edgecolor("black")
        arrow.set_linewidth(1)
plt.show()
plt.show()

# Text report showing the rules of a decision tree -

print(tree.export_text(best_model, feature_names=feature_names, show_weights=True))

|--- lead_time <= 150.50
|   |--- no_of_special_requests <= 0.50
|   |   |--- market_segment_type_Online <= 0.50
|   |   |   |--- weights: [3374.00, 351.00] class: 0
|   |   |--- market_segment_type_Online >  0.50
|   |   |   |--- weights: [3630.00, 3810.00] class: 1
|   |--- no_of_special_requests >  0.50
|   |   |--- weights: [11462.00, 2042.00] class: 0
|--- lead_time >  150.50
|   |--- weights: [1205.00, 3859.00] class: 1

# importance of features in the tree building ( The importance of a feature is computed as the
# (normalized) total reduction of the 'criterion' brought by that feature. It is also known as the Gini importance )

print(
    pd.DataFrame(
        best_model.feature_importances_, columns=["Imp"], index=X_train.columns
    ).sort_values(by="Imp", ascending=False)
)

                                           Imp
lead_time                             0.599031
market_segment_type_Online            0.237032
no_of_special_requests                0.163938
no_of_week_nights                     0.000000
required_car_parking_space            0.000000
market_segment_type_Offline           0.000000
market_segment_type_Corporate         0.000000
market_segment_type_Complementary     0.000000
room_type_reserved_Room_Type 7        0.000000
room_type_reserved_Room_Type 6        0.000000
room_type_reserved_Room_Type 5        0.000000
room_type_reserved_Room_Type 4        0.000000
room_type_reserved_Room_Type 3        0.000000
room_type_reserved_Room_Type 2        0.000000
type_of_meal_plan_Not Selected        0.000000
type_of_meal_plan_Meal Plan 3         0.000000
type_of_meal_plan_Meal Plan 2         0.000000
no_of_children                        0.000000
avg_price_per_room                    0.000000
no_of_previous_bookings_not_canceled  0.000000
no_of_previous_cancellations          0.000000
repeated_guest                        0.000000
arrival_date                          0.000000
arrival_month                         0.000000
arrival_year                          0.000000
no_of_weekend_nights                  0.000000
no_of_adults                          0.000000

importances = best_model.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(12, 12))
plt.title("Feature Importances")
plt.barh(range(len(indices)), importances[indices], color="violet", align="center")
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel("Relative Importance")
plt.show()

Lead_time,market_segment_type_Online,no_of_Special_Requests are important features for post-pruning.

Comparing all the decision tree models

# training performance comparison

models_train_comp_df = pd.concat(
    [
        decision_tree_perf_train.T,
        decision_tree_tune_perf_train.T,
        decision_tree_postpruned_perf_train.T,
    ],
    axis=1,
)
models_train_comp_df.columns = [
    "Decision Tree sklearn",
    "Decision Tree (Pre-Pruning)",
    "Decision Tree (Post-Pruning)",
]
print("Training performance comparison:")
models_train_comp_df

Training performance comparison:

	Decision Tree sklearn	Decision Tree (Pre-Pruning)	Decision Tree (Post-Pruning)
Accuracy	0.996200	0.571621	0.756903
Recall	0.988770	0.966706	0.762175
Precision	1.000000	0.439559	0.613324
F1	0.994353	0.604330	0.679695

# test performance comparison

models_train_comp_df = pd.concat(
    [
        decision_tree_perf_test.T,
        decision_tree_tune_perf_test.T,
        decision_tree_postpruned_perf_test.T,
    ],
    axis=1,
)
models_train_comp_df.columns = [
    "Decision Tree sklearn",
    "Decision Tree (Pre-Pruning)",
    "Decision Tree (Post-Pruning)",
]
print("Test set performance comparison:")
models_train_comp_df

Test set performance comparison:

	Decision Tree sklearn	Decision Tree (Pre-Pruning)	Decision Tree (Post-Pruning)
Accuracy	0.793393	0.574231	0.755257
Recall	0.702690	0.967217	0.766900
Precision	0.702214	0.447536	0.618978
F1	0.702452	0.611930	0.685045

Conclusion

• Decision tree model with pre-pruning has given the best recall score 98.87% on training data.
• The pre-pruned and the post-pruned models have reduced overfitting of the model.
• The Pre-Puned model gives a better performance on training and test sets . Hence we can use this model for decion making.

Actionable Insights and Recommendations

• Since repeated guests have very less chance of cancelling the booking, send Special Offers to the email of Special Guests so that they book the same Star hotel again.
• As the guests that make Special Requests show less chances of cancellation, provide the options of Special Requests for all the guests at the time of booking. Provide some complementary offers alongside that they will prefer.
• Provide Special Offers and Rates to Customers who book for long duration that involves more than 2 weekend nights or more than 5 weekdays. Offers such as one weekday/weekend night free with 2 weekend nights/5 weekday nights.
• Provide free Car parking space as Special Offers for few guests and see if that improves the Non Cancellation.
• Regularly send remainder mails to customers with long lead time/ arrival year. Ask if they would like to keep or cancel their booking. So that Hotel can get a status sooner.
• Provide Special offers for Summer months.
• Make some complementary offers for rooms with High Average Price.
• Improve the dishes provided in Meal Plan 2 which has higher cancellations.
• Redesign Room Type 6 which has higher cancellations.

Let's have a quick final overview of the data using the pandas profiling library.

# import pandas_profiling
from pandas_profiling import ProfileReport

# Use the original dataframe, so that original features are considered
prof = ProfileReport(data)
# to view report created by pandas profile
prof